Atari and Microsoft and the ST

[Petran79]

Amiga and atarist were both popular. Probably you are right about the seconds popularity.Amiga had more highend models. Eg i played games on an expensive amiga 2000 (2000 dm the cost back then).

Pc shops sold games and accessories for both platforms anyway.

 

I mean that in the end, users are rather conservative and resistant to change. My parents used a win 3.1 486 pc till early 2000. Due to word compatibility and lack of web they had to upgrade in the end.for the vast majority who type, chat, watch video etc all these new gadgets like smartphones are redundant, yet in a few years they'll have to adapt to them

[kool kitty89]

Amiga and atarist were both popular. Probably you are right about the seconds popularity.Amiga had more highend models. Eg i played games on an expensive amiga 2000 (2000 dm the cost back then).

Pc shops sold games and accessories for both platforms anyway.

It was my impression that the Amiga gained the lead in popularity around 1990 (it had started changing in '88/89 when the Amiga 500 matched the 520STFM in price, at least temporarily), and it was also around that time that heavily Amiga-specific games started becoming more common. (more games that were Amiga first and ST second, often with rather poor conversions to the ST -Shadow of the Beast is a prime example)

 

I mean that in the end, users are rather conservative and resistant to change. My parents used a win 3.1 486 pc till early 2000. Due to word compatibility and lack of web they had to upgrade in the end.for the vast majority who type, chat, watch video etc all these new gadgets like smartphones are redundant, yet in a few years they'll have to adapt to them

Sometimes it's good to stick with older stuff too . . . newer operating systems are often worse (unstable/buggy) early on and almost always use a lot more resource (especially memory) compared to older ones. (for a lot of things, Windows XP is still more efficient to use than Vista or 7 32 or 64-bit, Windows 2000 is still pretty useful too . . . hell, I know some people with rather nice Win98SE set-ups even -with a lot of 3rd party patches to keep up after MS dropped support)

Existing OSs would probably last a lot longer if not for declining support.

 

There's also often little need to upgrade if your old system does everything you need it to do. (my grandpa is still using his old 733 MHz Pentium III system, mainly for checking email . . . he'd probably still be running Win98 if not for the lack of update support, albeit he also may have ended up getting a new computer if not for my dad providing tech support -especially when they installed Service Pack 2 on that 256 MB system, and it was RDRAM based at that, so it was a bit of a pain to find a cheap upgrade option -in the end, my dad managed to find a good deal on a 512M RDRAM stick and bumped the system to 640 MB)

Our shared home/family computer is an "old" Athlon XP 3200 (might be a Sempron, I forget) system with 1 GB of DDR. (until I switched to my laptop 2 years ago, I was running an Athlon 1600+ with 512 MB, and it was fine for anything but newer games)

We used to have more regularly upgraded machines (I got that athlon set-up around 2004), but even then there were things like using Windows 98 until about 2003 (save my dad's office machine running 2000), though I seem to remember upgrading from DOS/3.1 to 95 and then 95 to 98/98SE relatively soon after the new releases.

 

Obviously the cases with my family's stuff isn't as extreme as 486+3.1 in 2000, though my grandpa's 733 MHz PIII machine still in use today is somewhat close to that. (more like if that 486 system had been upgraded with win9x and enough RAM to make that usable)

 

 

 

Anyway, that's still not really contradictory with companies pushing ahead. If the older/cheaper machines were still selling, they'd tend to remain on the market (look at the Spectrum and C64 . . . or the plain 520 ST/STF for that matter -in the PC world, the baseline 8088 based Tandy 1000 lasted to the end of the 80s I think, and in Japan you had the MSX and PC8801 lasting to the early 90s), but that doesn't mean those companies shouldn't be pushing for newer and more high-end machines as well. (that was a major problem with the ST and Amiga, both took a long time to get any real upgrades to the base architecture, though higher-end versions of the same old hardware happened sooner -Amiga 2000 and Mega ST- though Atari then went a bit odd with the too-high-end TT that didn't fit most of their established market and wasn't an amazing value for the price, and they didn't offer a lower-end derivative of that either -the MEGA STE probably would have done better in place of the TT in 1990 -though that was a bit lacking it was at least more cost effective, or something like the MSTE with TT video -but they could/should have been doing better than that earlier on, not necessarily with super fancy Amiga-like hardware, but perhaps something to match VGA with similarly modest hardware acceleration and low-cost aim -but I went off on a tangent with that earlier )

Edited June 10, 2011 by kool kitty89

[atarian63]

Oky, oky, oky.....

 

By 1994, the ST and Amiga were footnotes. You may not like it (I didn't), but the PC ruled the roost by then.

 

The PC had fast 486 chips with large motherboard L2 caches (and the 1st batch of Pentiums), Vesa Local Bus graphics cards with crazy fast (for the time) VGA chipsets like the Tseng ET4000, and 16 bit Soundblaster cards.

Yes, and by then they were kicking the crap about of anything the Amiga could pull off. (aside from modified systems with an '040 or 060 and 3rd party video card . . . for the few games on the system that even supported that -with AGA, you were stuck with the overhead of packed to planar conversion and the huge bottleneck of the chipRAM bus, the old Amiga chipset was nothing but a liability by that point, unfortunately . . . well aside from comparing it to super cheap/crappy 386SX 16-bit ISA PCs of that period -an SX-33 should have been a good bit better though, at least better than an A1200, let alone one without fastRAM)

 

As for the MT-32... It wasn't a sound card, it was a midi module that happened to get adopted for PC uses. Until better wavetable cards came along, the MT-32 with a Soundblaster was some of the best audio you could get, Not to mention that the MT-32 cost almost no CPU or bus time

Yes, it was a high-end (but still consumer-oriented) synthesizer module. There was the fully compatible LAPC-I card released as well. (they never seemed to consolidate the chipset down to a more competitive lower-cost level, let alone adding DMA sound support or updating the instrument sample set -an embedded MT-32 with soundblaster compatible PCM and gameport could have been an ideal higher-end single card sound solution even into the early 90s -longer if the standard persisted, albeit SB compatibility wasn't even that important if they added PCM support early enough as games would tend to support it specifically . . . from the actual synth side of things, allowing actual PCM samples to be imported and used as attack or percussion samples in addition to the default ROM ones would have been great)

 

OTOH, the MT-32 would also have seemed less amazing if more developers actually put effort into good Adlib/SB sound engines (pretty much no games used any samples for music in spite of that nice DMA sound channel the SB offered, and FM usage was pretty mediocre much of the time). The OPL3 was also sorely underutilized in spite of Soundblaster being heavily supported through the mid 90s. (several later gen games didn't even support the OPL3 or dual OPL2s of the Pro 1.0, and only gave token support to the old OPL2 . . . interestingly, it was often developers who continued to include MT-32 support that also had better OPL3 support -like Lucas Arts- while the likes of id never did support the MT-32 and never supported the OPL3 for that matter -Doom doesn't even have an option for SB Pro or SB 16 for that matter, it seems to only cater to the old 8-bit mono PCM and OPL2 synth)

 

The GUS: The Gus was a nice try, but by the time gravis had refined the cards enough to be usable, the Soundblaster AWE32 was out. The original GUS was very difficult to configure and get working with PC games.

That, and you had software MIDI drivers appearing for the SB/SB-16 too, including the default MS Windows General Midi driver. (albeit that didn't become a really attractive option until faster CPUs became common)

We never ended up using higher end midi support for those games though, pretty much everything was configured for SB/SB-16 regardless. (and an AWE-32 would actually be detrimental to that due to the greater amount of noise/distortion and some flaws with the CQM OPL3 clone used -the AWE-64 still has the clone flaws, but very clean sound and very little filtering -it sounds more like DOSBox)

 

Actually, I don't remember using any general midi (or other sample synth formats) aside from Sonic and Knuckled collection in the late 90s (I'm pretty sure we used the standard Windows software driver for that as it was definitely the Roland sound font and I know we didn't have a Sound Canvas or Roland Midi card).

 

 

 

 

 

Thanks for the education on Cache but I guess what I meant to say is cache always helped to when memory was expensive already and high speed memory was even more so a bit of high speed cache memory usually gave a nice performance bonus.

Yes, caching is always important, then and now . . . for any platform that can't run in main RAM at full speed all the time.

 

It's not just about having fast RAM either, DRAM simply can't be consistently fast like that for random accesses, any system much beyond an 8 MHz 286 probably would be suffering significant wait states in even with memory also clocked at that speed. (random accesses are always slow, but early CPUs were slow enough that was was often a non-issue, that didn't stay true later on and random access speeds crept up very slowly while CPU speeds increased dramatically; Fast Page Mode DRAM offered much faster burst accesses for reads/writes within the same row of DRAM cells, but still low accesses whenever you changed "pages" -so not only did caching become important to run at full-speed off the bus, but also made use of page mode accesses much more efficient by maximizing burst accesses)

With a 25 MHz 386 (be it SX or DX), you only needed 12.5 MHz (80 ns)FPM DRAM to allow full bandwidth in page mode (the 386 takes 2 cycles to complete an access), but you'd incur 3 added wait states for every page break (5 cycles per access rather than 2)

 

And again, the on-chip 32-bit L1 cache of the Cyrix and IBM 386/486 SLCs were better than just having the external 16-bit caching (though a 386DX wouldn't be much worse off with external caching, or no worse off at all if that cache RAM was at full 386 bus speed), though having external caching in addition to the on-chip cache would be better still. (still worse off than a full 386 of the same clock speed and caching though)

 

As to the ST/Amiga at least in the USA by 1993 or so most people had either left those two consoles for the newest dedicated game machines or the latest 486/Pentium technology. With the latest bus tech, new processors, video cards, sound cards the PC gaming experience was becoming quite good and a lot of publishers were leaving the other computers (amiga/st, old 8bits) to write games for the new PC computers.

It took longer for PCs to really catch on in Europe, but the game console comment is definitely accurate in that region.

 

As for Windows both 3.x and 95, you still could use Dos boot disks to play the games without any Windows issues. The consoles in the mid 1990s at least in the US imo were lacking in good flight combat sims, RTS, and first person shooters. I was never big into mario/sonic games, fighting games like street fighter/mortal combat even though a lot of people enjoyed those games and felt they were great I just enjoyed different genre of games and the PC world had the most/best games IMO for that.

Exactly, windows (early on) was mainly a gateway to easier access to existing DOS software (easier to install, configure, run, etc), though a few multimedia games required windows to run. (mainly those using win media player or the early equivalent -video for windows)

 

Many games weren't even available for windows-specific versions until around 1997, though a lot of games supported hardware acceleration prior to that. (Quake, Tomb Raider, etc)

Windows games almost always had hardware acceleration support (I can't think of any that didn't -I think even Sonic CD supported 2D acceleration). Support for hardware acceleration was one of the main reasons for the shift to high-level/library based programming, though even then a lot of games had some low-level tweaks. (I think that tended to be more on the CPU end since that was relatively fixed -you could count on an x86 CPU obviously- whereas the graphics cards varied a lot -early on, games had to be re-compiled for different APIs as well, as certain cards didn't support the likes of DirectX or Open GL -those had yet to become the defacto standards, and some developers had their own custom APIs that were supported on certain cards)

 

Of course, that high-level programming also greatly facilitated ports to consoles (aside from those with poor high-level tool support like the Saturn and PS2 -the latter got a ton of support from 3rd parties pushing their own libraries though).

 

 

Stardust sounds like an insanely bad example of crappy programming. (it almost sounds like they used a full AGA software emulator to run it . . . after all that's what Sonic and Knuckles Collection did -aside from the sound, it's running a Genesis emulator, the main reason it required a fast Pentium to run well, though that was in the late 90s, so less of an issue anyway -the 1995 Sonic CD PC release was a full port OTOH and required much less resource)

Sadly PC's still could not play an arcade game worth a damn though, just awful at it.

[oky2000]

 

Sadly PC's still could not play an arcade game worth a damn though, just awful at it.

 

Spot on, to play Super Stardust in 94 all you needed was £299 Amiga 1200 or £1000+ PC (and 2 hours to make it work).

 

Doom games yes, 50fps smooth movement arcade no. Then again £399 Plystation game needed £1500 PC too. In the EU in 94 there was only PPC Mac and x86 PC computers for sale in shops.

[barnieg]

The cost of Amiga's v PC's played a big part in the success within the UK. Disposable income was lower in the UK so a top of the range PC would have been out of most peoples budgets. I'm sure in theory that games like Super Stardust could have been optimised for a PC but at that point you are getting into optimised "banging the metal" coding which would probably result in all sorts of compatibility problems so there was no guarantee that everyone had the same hardware.

 

I would guess that a lot of Amigas and STs were sold into homes and treated as games consoles particularly the all in one 530 stfms and Amiga 500s. This became harder to pull off once games started to need more than 1 disk.

 

Also in the UK the Amigas and STs didn't really fail because of the products but because of bad decisions by the parent companies (and to this day the Amiga has a sizeable niche market that is profitable for a small number of businesses)

 

I stuck with my Amiga 500 through college, later upgraded to an Amiga 4000 with RTG graphics and Accelerator and for a long time I was able to do an a lot with my set-up (and sometimes more than the PC equivalent)

 

What's been quite disappointing is the end of semi-professional/hobbyist software market - Software that started on the ST or Amiga such as DTP or 3d packages which when PC/MAC only were creeping towards 10x the cost - one of the reasons I prefer open source today.

 

Barnie

 

 

 

Sadly PC's still could not play an arcade game worth a damn though, just awful at it.

 

Spot on, to play Super Stardust in 94 all you needed was £299 Amiga 1200 or £1000+ PC (and 2 hours to make it work).

 

Doom games yes, 50fps smooth movement arcade no. Then again £399 Plystation game needed £1500 PC too. In the EU in 94 there was only PPC Mac and x86 PC computers for sale in shops.

[kool kitty89]

Sadly PC's still could not play an arcade game worth a damn though, just awful at it.

PCs have been good for current arcade games since the mid 90s at least (look at Sega's 3D arcade ports ), and prior to that it was up to the type of game and (more so) simply software support. The US computer game market was very niche (at least after the C64 craze died down), and PCs weren't popular in Europe, so (regardless of technical ability) the C64, Amiga, and ST (and Spectrum . . . and CPC to some extent) got much better support for games in general than PCs in the US. (in Japan, computers were just about as niche as the US -mainly MSX PC8801 and PC9801 -the latter 2 very PC like is weak graphics and empahsis on business applications, with niche genres -especially RPGs- rather like US computer games had with adventures, RPGs, and flight sims)

 

The Tandy 1000 would have been great for a lot of early and some mid 80s arcade games.

 

From the mid 90s onward though, PC hardware has basically been game console hardware (if you had an accelerated graphics card at least -which was the case for any gaming-oriented PC from about 1996 onward). Beyond that it's very similar for consoles and PCs with some games better on some platforms, different exclusives (can't play Zelda on PC, can't play Grim Fandago on consoles -could play FFVII, Tomb Raider, and many Sega titles on both), PCs having the need for installation (turns some people off), consoles having more local multiplayer support (though that's getting less and less now), etc.

[kool kitty89]

The cost of Amiga's v PC's played a big part in the success within the UK. Disposable income was lower in the UK so a top of the range PC would have been out of most peoples budgets. I'm sure in theory that games like Super Stardust could have been optimised for a PC but at that point you are getting into optimised "banging the metal" coding which would probably result in all sorts of compatibility problems so there was no guarantee that everyone had the same hardware.

And the sheer capabilities of PCs. Even had the likes of the Tandy 1000 been offered in Europe at fairly ST-like prices and with far better game supporting hardware (and standard PC-expansion), it wouldn't have been as good of a deal in general. (and only in the US would its PC compatibility be a massive boon to market interest)

 

But as it was, things were far worse than that in Europe with only very expensive or very poor cheap PCs available. (and the "cheap" machines weren't cheaper than the ST)

 

I would guess that a lot of Amigas and STs were sold into homes and treated as games consoles particularly the all in one 530 stfms and Amiga 500s. This became harder to pull off once games started to need more than 1 disk.

Yes, and the games were also attractive being cheaper than carts and/or offering more possibilities via high-capacity low-cost media.

 

In the US, PCs fell into a similar role in the mid 90s (when PC gaming went mainstream and was close to the popularity of console gaming for a time), though never as extreme as the European preference for computer games. (even the C64 in the mid 80s never got to the point of Spectrum/C64/ST/Amiga in Europe, and died in favor of the NES very fast in the late 80s)

 

Also in the UK the Amigas and STs didn't really fail because of the products but because of bad decisions by the parent companies (and to this day the Amiga has a sizeable niche market that is profitable for a small number of businesses)

Yes, very true, though those management decisions would extend to available products too. (both base computers and upgrades made available -and not just the lack of easy expansion on the ST or such, but lagging in promised peripherals too)

 

IBM did the same thing with various missed opportunities and mistakes with their products (PC hardware offerings -including missing out on getting into the "home" market early on with the botched PCjr) and marketing, and probably would have run the PC into the ground if not for the burst in clone manufacturers and a huge clone market becoming established rather quickly in the mid 80s and eventually surpassing IBM in standard setting -from Compaq introducing a 386 PC first to EISA beating MCA, to graphics accelerators, etc, etc. (even later, Intel would cease to be the sole standard for CPUs as 3rd parties introduced their own extensions to the architecture and eventually x86-64 in the extreme case of a major architectural upgrade to the entire line -prior to that you had competing hardware designs using similar ISAs -sometimes jumping ahead of Intel performance- with some special extensions like MMX, 3DNow!, SSE, etc, but never a total update to the basic architecture -including finally going beyond 8 registers with 16 featured in AMD's x86-64 . . . that's registers the programmer can see, the internal architecture of all x86 chips since the late 90s have been custom RISC cores with hardware micro-op translation logic to interpret x86 instructions)

 

IBM also missed out on positioning OS/2 as the standard successor to DOS when it was more capable and efficient than contemporary Windows incarnations. (and unlike DRI, IBM had the funding to push directly against MS . . . but MS managed to out-market them, out-price them, and get more OEMs to support win 3.x and then 9x than OS/2)

 

Hell, the Apple II ALMOST got to the PC's level of open market support, but was also driven into the ground in the end.

 

 

Atari didn't need to push super hard for advanced/cutting-edge hardware, but more moderate upgrades keeping with the original clean, cost-effective design. (faster CPUs, fastRAM CPU bus -or conversely, a dedicated video/coprocessor bus instead, sound upgrades, updated graphics possibly continuing with the CPU-heavy nature and having modest VGA-like acceleration or pushing for more hardware acceleration with forward compatibility and low-cost kept in mind) 16 MHz CPUs really should have been offered with the launch of the MEGA series (probably with fastRAM and/or dedicated graphics memory) and probably with an FPU option to cater to CAD and such. (16 MHz could later become the low-end standard with 020s and 030s being introduced higher-up), with separated CPU and video buses, the SHIFTER could already have enough bandwidth for 320x200x256 colors or 640x200x16 (or up to 384/768 by 224/256 NTSC/PAL if framebuffer addressing was extended to at least 96k) and productivity extended to 4 colors at least. (if the SHIFTER gained FPM RAM support, it could double that bandwidth and have VGA-level 640x480 70 Hz, still without using expensive RAM or more than a 16-bit bus -just rather cheap/common 120 ns FPM DRAM as the lynx was using and you'd have up to 16 MB/s)

Then things like a YM2203 on top of DMA sound, among others.

 

Lots more hardware possibilities, but I'll stop there. (or else I'll keep rambling further off topic )

 

I stuck with my Amiga 500 through college, later upgraded to an Amiga 4000 with RTG graphics and Accelerator and for a long time I was able to do an a lot with my set-up (and sometimes more than the PC equivalent)

Yes, plenty versatile machines to work with, and lots of capabilities, and if you were already an Amiga user, the 4000's backwards compatibility was a major consideration over a PC (even a good priced one), and the price advantage of PCs would also be less for new users as they'd have to start from scratch rather than building on existing systems. (though the best option would probably be a used machine -US uers had a LOT more options for that though, more so if you had the know-how to build a system from scratch or had a local dealer -or small time workshop/home business- with services to build/upgrade custom PCs at really good prices -and I can't imagine Europe having an infrastructure for that like the US had established by the early 90s -my dad most definitely knew how to build a machine, and he took the budget route with a mix of old parts/case/monitor and some cheap new components to build our home systems; most of those keyboards and cases are still around if not in use, and one or 2 of the monitors too -definitely one from the early 90s)

 

 

What's been quite disappointing is the end of semi-professional/hobbyist software market - Software that started on the ST or Amiga such as DTP or 3d packages which when PC/MAC only were creeping towards 10x the cost - one of the reasons I prefer open source today.

Yes, and that seems to have been a combination of business/marketing of updated hardware (and getting hardware out in a timely and cost-effective manner) as well as resulting decline in market share of those newer models. (the old models lasted longest for games/casual stuff while the "serious" applications seem to have declined more rapidly . . . though I'm going by anecdotes, discussions and such -and news articles/TV and retrospectives- as I wasn't around back then, let alone in Europe ;p -well I can go by my memories as a kid with computers in the early 90s and discussions with my dad -who was definitely involved in the UK/EU market in the mid/late 80s due to working at Metacomco -a lot of contract work on the ST and Amiga there, I think they did a fair amount for some Sinclair machines too, some ZX80/81/spectrum stuff and QL iirc, I forget if he mentioned if they did a significant amount of the conversion work for TOS . . . I seem to remember that they did Amiga BASIC and an ST BASIC that wasn't released, but also his mentioning something about working on the ST's OS as well)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sadly PC's still could not play an arcade game worth a damn though, just awful at it.

 

Spot on, to play Super Stardust in 94 all you needed was £299 Amiga 1200 or £1000+ PC (and 2 hours to make it work).

OK, now you're either being a total idiot or a troll, I'm sorry but I already addressed that and a lot of it is just common sense. (by the time of fast 286s and VGA, PC hardware was already getting close to matching the Amiga's graphics hardware on its terms, and beating it on the PC's terms -note, but "Amiga's terms" I don't be crappy ports of Amiga games to PC, but reasonably well-made games on both ends that cater to the Amiga's strengths -ie pure 2D in its color limitations, no realtime scaling, rotation, or 3D effects)

 

But by 1994, 386s were already in the low-end range and 286 machines weren't available new anymore. A 386SX-33 machine with embedded VGA graphics (or basic 256 MB 16-bit ISA VGA card) should have generally met or exceeded the Amiga 1200 in raw performance for its type of games (ie common MD/SNES style 2D scrolling+sprite games) while being well ahead in anything 3D/pseudo 3D related (due to VGA's packed pixels and the 33 MHz 386SX being faster at many things than the 14 MHz 020), and probably exceeded it in every performance category if that 1200 didn't have fastRAM.

Such a machine would have been aging out rather quickly (in terms of availability on the new market) too as the fast 386DX machines became the de-facto low-end segment for new computers (given your 1994 context), especially AM386DX40 based rigs. (and late model 386 motherboards also had VESA support) I think 386SX machines had already been phased out in many markets by that point, but I'm not positive. (and I know some of the enhanced SX derivatives -like Cyrix and IBM had- were being used into the mid 90s at least -and int he extreme case of IBM's 486DLC 66/33, it exceeded the performance of most slower 486 based systems)

 

Smarter and/or more capable users could aim more at the variety of used computers at good values available at the time and/or build their own system (or get a friend -or paid- assistance to build a nice custom PC for much less than any new pre-built systems on the market).

That's where you'd go if you wanted a machine in the same price range as a 1200. (more so if you already had an existing PC to build on -even if you had to junk the motherboard, accessory cards, power supply, case, and monitor could still be re-used -unless all of those were super out-dated)

Of course, if you were building an A1200 based system, you'd need to consider the costs of an RGB monitor (for any decent computing at least), fastRAM expansion, and other add-ons depending on your needs.

The pre-built PC market (with few exceptions) never did cater to high quality/performance machines at low/tight prices, cheap pre-built machines were often crap that was hardly worth the value, and the OK value machines were powerful but also expensive. (Tandy seems to have been among the exceptions, Atari perhaps as well -thouch their PC seem to have been sparsely distributed, but for the most part -especially in the early 90s up to today- you'd need to shop around for a good deal on a used machine and/or build/upgrade your own for far lower cost than anything close on the pre-built market -in recent years, there's been some decently priced complete desktops in the high $6-800 range, so definitely a lot better than things once were, but one could still shop around and build a similar machine for significantly less -or less still if they had an existing machine to build on-to -on another note, HDTVs have made it rather easy to use as PC monitors)

 

 

Of course, available software differs from raw performance, and while most PC games (especially high-profile ones) were pushing the hardware rather well, some sloppy ports didn't, and others weren't available. (European developers obviously had a strong bias towards the Amiga, and US Amiga Users benefited greatly from that too with many, many games they otherwise would never have had )

That issue also applies to the lack of AGA support in general for Amiga, which makes the A1200 rather moot in general . . . games that didn't need that CPU grunt rarely benefited from the A1200 at all, and games that did tended to fall well short of PC counterparts unless horribly coded on the PC -or making use of hefty accelerators on the Amiga that boosted their resource above that of common PCs. (which was largely crippled by the planar graphics, horribly slow chipRAM access speeds for the CPU, and only the CD32 had Akiko -which wasn't as good as direct packed pixel graphics and still had the chipRAM bottlenecks on top of that)

It took a while for VGA to become the de facto gaming standard following its 1987 introduction, but it had gained that status by 1990 (all major games supported it, and most catered to it specifically -usually with low quality support for Tandy, EGA, and VGA too, only a few shareware games continued to push plain EGA a bit longer -the same sort of games reluctant to even support Adlib; better games had already added MCGA/VGA compatibility modes by 1988 -especially necessary since MCGA had no EGA compatibility- but those very early examples just did it for compatibility and didn't take much advantage of MCGA/VGA's color, hardware scrolling, raster op acceleration, or packed pixel rendering advantages) . . . so it makes sense (given how late it was) that AGA wouldn't catch on quickly, especially with it being a far less significant upgrade.

 

 

 

Hmm, wait a minute: 1994, didn't you mention the PC version Stardust was a Win9x-specific game? (if it was '94, it should have been DOS-specific for sure, and would be catering to low-level programming . . . though many early windows games did that too into the late 90s . . . so it would have to be really poorly done to require a pentium)

And that would also make even less sense for it requiring SVGA.

Stardust is an extremely basic game (and extremely out of date for "arcade" in the context of 1994), a fixed-screen asteroids clone without scrolling and with relatively little on-screen . . . something that might even make it feasible on the bus-strangled CD-32 in 256 colors. (rather than dual playfield mode)

There should have been no need for anything above a bog-standard 256k VGA card for 320x240 in 256 colors with double buffering in VGA memory. (albeit most/all DOS and many windows games were still supporting 320x200 into the late 90s -Tomb Raider II did in 1997)

If you wanted current-gen arcade stuff (ie almost all 3D), the PC had a lot, especially in the mid/late 90s. (Sega did a lot of good PC games in that period)

 

On that note: windows 9x didn't make gaming easier/better for PC users because of windows games (that didn't happen until about 1997 with hardware acceleration being universally supported and APIs getting more efficient), but because windows allowed DOS programs to be installed/configured and run much more easily (even for the tech-capable, it made driver issues less problematic, especially stuff like extended memory management), and the vast majority of games were still DOS-only until 1996 (aside from a few needing video for windows) and not predominantly windows-specific until 1997.

I think OS/2 may also have greatly alleviated some of those problem earlier on (and Warp was launched in '94 too), but I'm not sure.

 

 

Now, if you wanted something as cheap and rougly as capable as the A1200, you'd be looking in the used market by that point

Doom games yes, 50fps smooth movement arcade no. Then again £399 Plystation game needed £1500 PC too. In the EU in 94 there was only PPC Mac and x86 PC computers for sale in shops.

Now you're really being silly. A game like Doom needs a lot of 2D (blitting) rendering muscle to run at reasonable speeds, and any machine capable of running it reasonably well would also be excellent for conventional 2D. (a machine running it barely adequately at lower detail would be limited for 2D though, but still possibly on par with the AGA amiga -definitely on par in some contexts, but playing on its terms with heavy use of sprites and dual 16 color playfields would be less favorable)

 

And it wouldn't be 50 FPS either, it would be 70 Hz, or drops thereof. (if a game couldn't manage 70, it would be 35 and still pretty good . . . though scrolling was often full 70 Hz, especially for single background plane stuff -where VGA's hardware scrolling was most useful . . . even those 8086 examples posted a while back were scrolling at 70 Hz, and even slow 8-bit ISA cards could do that -they were only limited for software rendered/copied graphics, not the VGA accelerated stuff)

 

 

 

That Playstation reference is just silly though, of course consoles will be cheaper, they're sold to a totally different market, and an increasingly cut-throat one at that. Computers and console prices were relatively close in the mid/late 80s due to a combination of very low-end computer hardware (much of which was designed as game console/arcade hardware and then worked into a home computer) being pushed at very tight margins while consoles were still being sold above cost and using expensive cartridge media that made them less cost competitive with computers.

Overall component prices fell due to competition and volume production that made comparable home computers less and less cost competitive (especially for the home-built machine -or shop-built machines at competitive prices, not so much for pre-built boxes though). The Amiga or ST probably could have lasted a lot longer (maybe persisted into a niche to this day -or continued as standard in Europe if it turned into a PC-like market with competitive clones and efficient 3rd party upgrades), but they didn't do that at all and both were nominally worse for up-to-date computing than PCs by the early 90s. (for games catering to the old 1985 hardware, the Amiga continued to be great into the early 90s, rather like the Megadrive -or C64 and NES slightly earlier)

 

With the coming of the Playstation, you had a very powerful system (comparable to some of the new 3D video cards being introduced in 1995) being sold far below cost for the first few years and using low-cost CD-ROM media (though prices on PC games were -and are- still generally lower than consoles, N64 was a much bigger gap on that end though). Had Sony sold the PSX's GPU (with onboard SGRAM) in 1995, it would probably have been priced significantly higher than the standalone playstation. (it would be at a profit, and not a slim profit either)

Plus, the PSX came just before PCs got that strong burst of custom graphics hardware that became standards quickly. (and by 1997, the PSX was aging and less comparable to new PCs . . . though a low-end PC with PSX-level graphics capabilities would be comparable; 1999/2000 would be more comparable with current gaming class PCs compared to the Dreamcast and PS2 -games were much cheaper for PCs than PS2 and lesser extent DC, though PCs were obviously more expensive due to the more flexible hardware and much, much higher margins, it was a very different market )

 

On top of that, a gaming PC is also a very flexible and useful tool as a general purpose machine. (in the context of the late 90s, in addition to games, you had general computer tasks for work/school, and non-game entertainment uses like a DVD player -and a PC based DVD player could save several hundred dollars over the cost of standalone players in those days -an average user would just need a reasonable late 90s CPU and something like a Rage Pro AGP card for DVD play back and composite and s-video output for a TV -in our case, by dad hacked some beta drivers to make it work with a PCI Rage Pro )

 

That same thing had continued for all subsequent consoles with the exception of Nintendo's carefully designed/configured machines sold closer to how things were in the 80s and early 90s, and even there it's at far lower margins than any computer market would push. (ie at cost for launch and later for profit . . . except for the Wii, which probably was at a very substantial profit from day 1 -probably over 100% at retail, though less for what Nintendo was selling it to distributors at . . . and in that case, the Wii is very overpriced tech sold at an affordable price point, sort of like the shitty low-end PCs available in the late 80s and early 90s)

 

 

 

 

 

 

 

Hmm, thinking on this topic in general: Europe had no-one like Tandy at all early on (for OK value PCs at reasonable mid/low consumer prices) and the PC market continued to lack a stronger brand-name/pre-built line of systems (and I already said it would have been interesting if Amstrad or the like had managed licensed distribution of that design), but that situation remained in the early 90s even with cost competitive off the shelf parts becoming widely available. So, a company could take the initiative to push off the shelf PC hardware at comparably tight margins to previous home computers and made the market far, far more open to many who couldn't/wouldn't find a good used system, build their own machine, or find a place configuring custom PCs at good prices.

From what I understand, competitively priced off the shelf PC motherboards were far too competitive to merit custom motherboards (even consolidated/non-expandable ones), so one would at least have to work off that angle, though, short of that, they might use a custom motherboard with off the shelf mass market embedded PC chipset instead. (otherwise it could be hard to really push something in a ST/Amiga console type form factor . . . for desktop boxes, off the shelf boards would probably be cheapest by far)

 

Atari could have been in a prime position to do that in the early 90s in Europe, though balancing that with their STs could be another matter. (and bad management contributing to the ST's decline would also tend to apply to any PC exploits . . . and if they'd managed the ST better -including hardware updates- they might have had a strong enough market to not even consider introducing PCs in Europe)

In the US, they'd probably have more of an issue with branding, but still would be in a virtually untapped merket had they pushed low-margin, low price, yet relatively capable PCs. (it would be interesting to find out the pricing on their existing PC line, especially since they seem to be significantly better than Commodore or Amstrad's offerings -the PC4 would have been a great upper mid-range machine towards the end of the 80s and a good lower end machine in the early 90s . . . except it seems to have been discontinued by 1989)

 

As far as a game-oriented PC would go, a 16 MHz 286 with 640k and 256kB VGA card (or embedded video -as Atari machines had) and a soundblaster would have been good for pretty much any game up through 1992 at reasonable/playable speed/quality (a few games might run too fast), and should have been better than the Amiga 500 for many things. (though you couldn't use it on a TV)

 

However, if Atari had been doing reasonable updates to the ST, they should have already had something with VGA-like capabilities, 16 MHz 68000, and updated sound by the end of the 80s, and if they HAD had that in Europe, such a 286 PC would probably be unnecessary and both less capable and more costly than a clean evolution of the ST. (let alone potential higher-end models and the fact Atari's machine could still be catering to TVs -and at that point in the late 80s and very early 90s, the amount of silicon going to maintaining 100% PC compatibility and MDA/CGA/EGA compatibility for VGA -as well as VGA-specific text modes- would be significantly more cumbersome than the relatively clean design of the ST -especially pre TT, even more so pre-blitter- and it wouldn't be until later that such embedded silicon would become inconsequential to overall cost)

 

 

In the US, both the ST and Amiga were screwed by the end of the 80s, but even with good, regular hardware updates and continually good management, they would have likely only stayed attractive (for prospective buyers of computer games) until about 1996 when PCs really began getting hardware acceleration support and easier to use interfaces and the hugely competitive consoles took up the rest of the market. (Europe would have more of a chance due to building onto a strong userbase of compatible machines) Had the ST stayed more cost effective than contemporary PCs with reasonably up to day feature sets, and managed PC file compatibility like the Mac ended up with, it might have maintained a niche to this day. (though maintained Euro-support would have been the backbone of the line in any case)

Edited June 14, 2011 by kool kitty89

[ApolloBoy]

MT-32 = No sound effects, just instruments so like a cheap ass keyboard not a Fairlight sampling keyboard as is the case with Amiga sound hardware. That was my point, the Gravis Ultrasound pissed all over limited and blinkered rubbish like MT-32. You could do the same music AND multiple 16bit sample playback. If you had MT-32 it was like the SNES soundchip, basically like you have to use existing instruments not sample your own

As a synth player, I did a double take when I read this. The MT-32 is actually a full-blown synth based on the Roland D-50, something that was used by loads of professional musicians and wasn't "limited and blinkered rubbish". Yes, it doesn't handle sound effects but it's nothing like the SNES.

[oky2000]

MT-32 = No sound effects, just instruments so like a cheap ass keyboard not a Fairlight sampling keyboard as is the case with Amiga sound hardware. That was my point, the Gravis Ultrasound pissed all over limited and blinkered rubbish like MT-32. You could do the same music AND multiple 16bit sample playback. If you had MT-32 it was like the SNES soundchip, basically like you have to use existing instruments not sample your own

As a synth player, I did a double take when I read this. The MT-32 is actually a full-blown synth based on the Roland D-50, something that was used by loads of professional musicians and wasn't "limited and blinkered rubbish". Yes, it doesn't handle sound effects but it's nothing like the SNES.

 

£400 for something that is only HALF the sound system of a home computer system for playing games on IS blinkered and inneffective rubbish indicative of what PC users think is a good idea and the rest of the sane world believe is a load of old tat. What's the point of a £350-400 sound card that ONLY plays music? Like I said it can't even produce music as good as Shadow of the Beast OR Agony on a stock £199 Amiga in 1992 because you are limited to instruments provided (which is where the comment about the SNES soundchip comes in) 100% FAIL

[ApolloBoy]

MT-32 = No sound effects, just instruments so like a cheap ass keyboard not a Fairlight sampling keyboard as is the case with Amiga sound hardware. That was my point, the Gravis Ultrasound pissed all over limited and blinkered rubbish like MT-32. You could do the same music AND multiple 16bit sample playback. If you had MT-32 it was like the SNES soundchip, basically like you have to use existing instruments not sample your own

As a synth player, I did a double take when I read this. The MT-32 is actually a full-blown synth based on the Roland D-50, something that was used by loads of professional musicians and wasn't "limited and blinkered rubbish". Yes, it doesn't handle sound effects but it's nothing like the SNES.

 

£400 for something that is only HALF the sound system of a home computer system for playing games on IS blinkered and inneffective rubbish indicative of what PC users think is a good idea and the rest of the sane world believe is a load of old tat. What's the point of a £350-400 sound card that ONLY plays music? Like I said it can't even produce music as good as Shadow of the Beast OR Agony on a stock £199 Amiga in 1992 because you are limited to instruments provided (which is where the comment about the SNES soundchip comes in) 100% FAIL

But like I said, the MT-32 is a full-blown synth, meaning that you can connect a MIDI keyboard to it and play it as if it were a D-50 or the like. You can even program it like a synth, albeit through MIDI (but many other synth modules were the same way around 1987). So it wasn't just sold as a PC add-on (in which case you were limited to the patches that each game used), it was an actual synth too. I've noodled around with a D-50, the MT-32's bigger brother, and it's nothing like the Nintendo S-DSP. The MT-32 (and every other synth that uses Linear Arithmetic synthesis) only uses samples for part of the sound while the rest is synthesized. The S-DSP on the other hand, is entirely sample-based so your comparison doesn't work. I'm sure kool_kitty would be able to fill in on this too.

[svenski]

US uers had a LOT more options for that though, more so if you had the know-how to build a system from scratch or had a local dealer -or small time workshop/home business- with services to build/upgrade custom PCs at really good prices -and I can't imagine Europe having an infrastructure for that like the US had established by the early 90s

 

*Sigh*, as someone who worked for a PC mail order business from the early to late nineties , I find the above somewhat condescending and similar to your previous remarks along the lines of Ronald Reagan arrived in London on a state visit and introduced Europe to disk drives....

[krupkaj]

I played Space Quest from Sierra online with MT-32. And I have to write it is really great. I even tried to program some sounds but it is not so easy. But there are sound banks available on the internet for experimenting.

[oky2000]

Can the MT-32 replicate, let alone improve qualitatively, the soundtrack to the 1994 game Super Stardust on a humble £300 Amiga? No.

 

I don't care if it has 48khz 16bit general MIDI type library of trumpets and what not, still a restrictive piece of shit. Ditto 1986 Ensoniq Mirage vs D50 ditto locked fixed instrument bank of SNES vs awesome Sega Gauntlet IV soundtrack.

 

MT32/LAPC1/SOUNBLASTER....all a load of overpriced bollox. Before Gravis or AWE32 it was all for losers kiddingthemselves their PC was worth £400 let alone the £2000 they wasted on it to play Doom losers.

[ApolloBoy]

I don't care if it has 48khz 16bit general MIDI type library of trumpets and what not, still a restrictive piece of shit. Ditto 1986 Ensoniq Mirage vs D50 ditto locked fixed instrument bank of SNES vs awesome Sega Gauntlet IV soundtrack.

Except the SNES doesn't have fixed instruments, it uses samples loaded from the game ROM, kinda like the Amiga. Also, the Ensoniq Mirage may have been a sampler, but it was low-res and was a bit restrictive compared to later samplers. The D-50, on the other hand, was a fully programmable synthesizer; you're comparing apples to oranges here.

 

MT32/LAPC1/SOUNBLASTER....all a load of overpriced bollox. Before Gravis or AWE32 it was all for losers kiddingthemselves their PC was worth £400 let alone the £2000 they wasted on it to play Doom losers.

Well if this is the attitude you're going to take, I'm not going to bother with you anymore. This is 2011, not 1994. *clicks Ignore User button*

 

kool_kitty, he's all yours.

[Pilsner73]

Can the MT-32 replicate, let alone improve qualitatively, the soundtrack to the 1994 game Super Stardust on a humble £300 Amiga? No.

 

I don't care if it has 48khz 16bit general MIDI type library of trumpets and what not, still a restrictive piece of shit. Ditto 1986 Ensoniq Mirage vs D50 ditto locked fixed instrument bank of SNES vs awesome Sega Gauntlet IV soundtrack.

 

MT32/LAPC1/SOUNBLASTER....all a load of overpriced bollox. Before Gravis or AWE32 it was all for losers kiddingthemselves their PC was worth £400 let alone the £2000 they wasted on it to play Doom losers.

 

The Commodore Amiga sound chip was great for the mid 80s but I really think the Amiga was starting to show it's age by the mid 90s. Didn't commodore go bankrupt around 1994 or so? Yeah the people adopting PC over the amiga in the mid 90s for games and productivity software were the losers.

 

The Amiga was a great system but thanks to Commodore sitting on the tech for so long, by 1995 PC computers and video game consoles were finally surpassing the Amiga. What IF commodore had not gone bankrupt and upgraded the Amiga chipset who knows, but Commodore went under and the Amiga lost what major software house support it had. Atari with the Falcon 030 was too late to the game also though the Falcon030 was a major upgrade of the ST tech just a bit late to the party.

Edited June 15, 2011 by Pilsner73

[Petran79]

Can the MT-32 replicate, let alone improve qualitatively, the soundtrack to the 1994 game Super Stardust on a humble £300 Amiga? No.

 

I don't care if it has 48khz 16bit general MIDI type library of trumpets and what not, still a restrictive piece of shit. Ditto 1986 Ensoniq Mirage vs D50 ditto locked fixed instrument bank of SNES vs awesome Sega Gauntlet IV soundtrack.

 

MT32/LAPC1/SOUNBLASTER....all a load of overpriced bollox. Before Gravis or AWE32 it was all for losers kiddingthemselves their PC was worth £400 let alone the £2000 they wasted on it to play Doom losers.

 

After cd-rom games became the standard this was not such a big issue. Besides, we did not pay much attention to sound differences.

some games used digitized music or mod files. There certainly was a quality difference in noise between an original sb16 and any clone.

But usually the sound speakers were not of the best quality either.

At that time those who could not afford a computer or console had a cheaper and better alternatives:arcades, often with no sound at all.

[+poobah]

Can the MT-32 replicate, let alone improve qualitatively, the soundtrack to the 1994 game Super Stardust on a humble £300 Amiga? No.

 

I don't care if it has 48khz 16bit general MIDI type library of trumpets and what not, still a restrictive piece of shit. Ditto 1986 Ensoniq Mirage vs D50 ditto locked fixed instrument bank of SNES vs awesome Sega Gauntlet IV soundtrack.

 

MT32/LAPC1/SOUNBLASTER....all a load of overpriced bollox. Before Gravis or AWE32 it was all for losers kiddingthemselves their PC was worth £400 let alone the £2000 they wasted on it to play Doom losers.

 

Blah blah blah.

 

How many musicians in the early 90s had an MT-32 or D-50.... Ok, now how many used an Amiga? Right.

 

The MT-32 is a midi synthesizer that happened to gain popularity as a sound module for computers (usually in conjunction with a sample based sound card).

 

In the early 90's, there was *NO BETTER* consumer level PC sound combo than a Soundblaster16 and an MT-32 until the Roland Sound Canvas came around.

[Petran79]

I remember various pc tests of the 90s. The creative soundblaster 16 scored quite well in signal to noise ratio, actually higher than the gravis ultrasound.

There war another test with wavetable cards. Roland sc had the best quality.there was also a sample cd of the cards music. Gamers would opt for the wave blaster daughterboard option for the sb16. Though that was short lived (cd music games were trend), but quite affordable in comparison to buying a sb awe 32 or a roland.