F18A programming, info, and resources

[+OLD CS1]

Besides that, the quality will never be as good. You are still starting with a composite (or, at best, component) video signal and upscaling it, as opposed to the F18A generating the VGA natively. That was what ultimately led to me abandoning that concept as well.

 

You'll find that those 15Khz RGB monitors will be harder and harder to find in working order, and the availability of those VDPs is not good enough to be any help to people who don't already have them. F18A is a good forward-looking solution, not just for the TI, but for all the systems that use the TMS9918A.

 

Meh, the quality is "good enough," but the price is comparable to an F18A, anyway. If your only purpose is a 9918A-based device, then the F18A is the way to go as Matt can put them together. Otherwise, you can use the converter to connect any composite and S-Video output device, as I am doing with my 64C, Atari 130XE, and the Jaxx Pacific hand-held arcade games. I have yet to try with my Sega Nomad, but I suspect it will work just as well if not better (it has a pretty good video output.)

[RXB]

Besides that, the quality will never be as good. You are still starting with a composite (or, at best, component) video signal and upscaling it, as opposed to the F18A generating the VGA natively. That was what ultimately led to me abandoning that concept as well.

 

You'll find that those 15Khz RGB monitors will be harder and harder to find in working order, and the availability of those VDPs is not good enough to be any help to people who don't already have them. F18A is a good forward-looking solution, not just for the TI, but for all the systems that use the TMS9918A.

Yea RGB monitors are still being made today.

 

The color is just so much better then what VGA can to do. VGA was designed for TEXT while RGB was designed for GRAPHICS and COLORS.

 

http://www.dpreview.com/news/2013/2/14/Dell-releases-UltraSharp-monitors-with-99percent-Adobe-RGB-color-space-support

 

 

VGA is just cheaper for a good reason.

Edited October 6, 2013 by RXB

[slinkeey]

So you are going to tell me that the composite video comming out of a standard TI-99/4A is better than ones with the F18A that uses VGA?

 

VGA has the following discreet signals.

Red, Green, Blue, Horizontal Sync, and Vertical Sync

[Omega-TI]

So you are going to tell me that the composite video comming out of a standard TI-99/4A is better than ones with the F18A that uses VGA?

 

VGA has the following discreet signals.

Red, Green, Blue, Horizontal Sync, and Vertical Sync

 

I'm with you on this one, we must be misunderstanding what he is saying. I've always understood that VGA stood Video GRAPHICS Array, nothing to do with text really. Now some of the early computers had a "Page White Display", but it was more of a marketing thing to do with the resolution (aimed at the word processing market) and of course the connector, which for a VGA is a 15 pin arrangement. VGA in the early days was defined as a minimum of 256 colors, where EGA was 16 simultaneous colors and CGA was just crap and not worth mentioning. Later came SVGA and others. I sure do not miss the era of those old monochrome jobs which WERE for text, and they came in three flavors, B&W, GREEN and AMBER, which always looked orange to me.

[Tursi]

Yea RGB monitors are still being made today.

The color is just so much better then what VGA can to do. VGA was designed for TEXT while RGB was designed for GRAPHICS and COLORS.

http://www.dpreview.com/news/2013/2/14/Dell-releases-UltraSharp-monitors-with-99percent-Adobe-RGB-color-space-support

VGA is just cheaper for a good reason.

You linked to an HDMI monitor. It doesn't support 15khz RGB.

 

This is just becoming a "yes-it-is-no-it-isn't" debate, no point continuing it.

[matthew180]

Yea RGB monitors are still being made today.

Really? So you can get a *new* fixed frequency 15KHz RGB monitor that uses a CRT and comes in a consumer friendly case (not just a chassis)? And for less than $350? Where? I know a lot of coin-op people who would like to talk to you.

 

The color is just so much better then what VGA can to do. VGA was designed for TEXT while RGB was designed for GRAPHICS and COLORS.

Huh?? You have no idea what you are saying. "RGB" means "red, green, blue", and *every* color TV or monitor made since the invention of color TV is RGB.

 

When we use the term "RGB" here, i.e. for classic home computers or arcade coin-op games, etc. it is generally accepted that "RGB" is referring to a CRT-based monitor or bare chassis that has a fixed horizontal frequency (usually 15KHz) and accepts separate red, green, and blue inputs as analogue voltages between 0.0 and 0.7V. It is this lower 15KHz fixed frequency that typically causes most of the problems when trying to interface newer monitors.

 

VGA means "Video Graphics Array" and was invented by IBM in the 80's and actually refers to the chipset on the graphics card. In typical fashion, industry and media muddled the term and now "VGA" is generally accepted to refer to a physical interface (15-pin high density sub-D connector) and video specification that includes resolution and refresh rates. A very common "VGA" resolution and refresh rate are 640x480@31KHz, and is typically what people are talking about here when they say "VGA".

 

VGA has the *SAME* analogue 0.0 to 0.7V inputs for red, green, and blue as an "RGB" monitor, and can therefore represent the *SAME* color depth as an "RGB" monitor. However, a "VGA" monitor can display higher resolutions and therefore create better graphics. The higher resolutions of VGA also lends to better rendering of text.

 

VGA is just cheaper for a good reason.

Uh, how about supply and demand.

[RXB]

You linked to an HDMI monitor. It doesn't support 15khz RGB.

 

This is just becoming a "yes-it-is-no-it-isn't" debate, no point continuing it.

I was not talking about the plug or the interface video card in the computer, but the monitor.

If you looked at the specs you would see that the monitor has a built in processor for RGB Adobe to correct and calibrate colors for Engineering and Graphics work.

You can not put this tech into the VIDEO CARD for output, it is in the MONITOR thus you misunderstand what I am talking about.

 

Also Kevan got it. RGB is still around today for that reason. If you knew the history of RGB and VGA it would make more sense to you.

 

https://www.google.com/search?q=RGB+Adobe&es_sm=93&biw=1642&bih=905&tbm=isch&tbo=u&source=univ&sa=X&ei=eh1TUu2KOe_wyAHGv4GQAg&ved=0CEUQsAQ

 

There are even arguments over sRGB vs RGB Adobe and a now a new standard in the EU is being discussed and designed.

 

http://fstoppers.com/adobergb-vs-srgb

 

 

The real issue is in a Lab with 40 people that use different computers and different video cards consistency is not possible with normal VGA monitors.

Monitors with built in calibrations using built in processors for this are miles ahead of VGA. (in other words no built in sRGB or RGB Adobe)

 

 

http://compreviews.about.com/od/monitors/tp/GraphicsLCDs.htm

(Low end monitors with reasonable output and these are not highend with sRGB or RGB Adobe)

 

Sorry but I am a new tech guy and keep up on things like this stuff. I doubt anyone in the TI community uses sRGB or RGB Adobe or works in a lab.

 

P.S. I worked IT for Freightliner and some times Customers would come in with a color they wanted on the Truck and we had to duplicate on the computer for mixing the paints.

Edited October 7, 2013 by RXB

[+InsaneMultitasker]

I was not talking about the plug or the interface video card in the computer, but the monitor.

If you looked at the specs you would see that the monitor has a built in processor for RGB Adobe to correct and calibrate colors for Engineering and Graphics work.

 

Did you post your reply in the wrong forum???

[matthew180]

I was not talking about the plug or the interface video card in the computer, but the monitor.

But that's mostly what "VGA", as a general term, means. You keep using the term incorrectly, or have some idea that VGA is something that is it not. VGA == analogue RGB when it comes to color space.

 

If you looked at the specs you would see that the monitor has a built in processor for RGB Adobe to correct and calibrate colors for Engineering and Graphics work.

And...? What does Adobe color correction have to do with a 30 year old 15KHz fixed frequency monitor's capability, or that the F18A outputs an analogue RGB color signal with a physical and electrical standard typically known as "VGA"?

 

You can not put this tech into the VIDEO CARD for output, it is in the MONITOR...

Uh, wrong. A video card can be calibrated with a monitor and people have been doing it for a long time. Typically only those working in a print or similar industry care because they want the colors on the screen to match the colors on the printed page. Calibration is required because monitors use red, green, and blue *light*, and printing uses cyan, magenta, yellow, and black *ink*. If Adobe came up with some chips that put the color calibration into the monitor, great, but what does that have to do with anything we are talking about? All monitors are still using red, green, and blue LEDs or phosphor (in the case of a CRT) which means the color space is still the same as it always has been.

 

... thus you misunderstand what I am talking about.

Then what are you talking about? I don't see how this has anything to do with the F18A or that despite the use of the standard "VGA" connector and electrical specification, the F18A colors are just as good (or better) than what you get on your fixed frequency 15KHz "RGB" monitor.

[TheMole]

I was not talking about the plug or the interface video card in the computer, but the monitor.

If you looked at the specs you would see that the monitor has a built in processor for RGB Adobe to correct and calibrate colors for Engineering and Graphics work.

You can not put this tech into the VIDEO CARD for output, it is in the MONITOR thus you misunderstand what I am talking about.

 

Also Kevan got it. RGB is still around today for that reason. If you knew the history of RGB and VGA it would make more sense to you.

 

https://www.google.com/search?q=RGB+Adobe&es_sm=93&biw=1642&bih=905&tbm=isch&tbo=u&source=univ&sa=X&ei=eh1TUu2KOe_wyAHGv4GQAg&ved=0CEUQsAQ

 

There are even arguments over sRGB vs RGB Adobe and a now a new standard in the EU is being discussed and designed.

 

http://fstoppers.com/adobergb-vs-srgb

 

 

The real issue is in a Lab with 40 people that use different computers and different video cards consistency is not possible with normal VGA monitors.

Monitors with built in calibrations using built in processors for this are miles ahead of VGA. (in other words no built in sRGB or RGB Adobe)

 

 

http://compreviews.about.com/od/monitors/tp/GraphicsLCDs.htm

(Low end monitors with reasonable output and these are not highend with sRGB or RGB Adobe)

 

Sorry but I am a new tech guy and keep up on things like this stuff. I doubt anyone in the TI community uses sRGB or RGB Adobe or works in a lab.

 

P.S. I worked IT for Freightliner and some times Customers would come in with a color they wanted on the Truck and we had to duplicate on the computer for mixing the paints.

 

Rich, you really need to be careful with these kinds of posts. For someone who doesn't know what all of these acronyms and technologies are about, your post could be construed to mean that you can buy the monitors you are referring to and use them with an unmodified TI, which is obviously not the case. If you want to use a modern "VGA" monitor (meaning a monitor with that typical blue 15 pin connector, called a DE-15 connector), you will need either a composite-to-VGA box or an F18A. If you use a composite-to-vga box, you will need to be aware of the fact that your image quality will, by definition, be at best on par with what you can get on composite and in reality quite a bit less than that due to the conversion steps necessary. The F18A on the other hand will give you a VGA quality signal out of the box, which is better than composite in any way measurable and on top of that doesn't need the conversion.

 

So no matter how you slice or dice it, objectively speaking the F18A is the best option for high quality video output from the TI, although subjectively other options might be good enough for some.

Edited October 8, 2013 by TheMole

[RXB]

Having worked on this problem in real life I have a different perspective on this.

I only mentioned it as people did not know that RGB still exists and just got rolled into other tech like sRGB or Adobe RGB.

 

Also what the computer says is the color in memory can be entirely off from what you see on the monitor this is the reason for adjustments.

 

Secondly I have first hand experience with holding something up to the screen and what the computer says the color is and what I am holding in my hand are not the same.

Physically the color is one thing but the monitor is not even close according to the computer it is the same color? Many hours wasted to fix this issue.

 

Yes the F18 is what some people want so fine. I was just pointing out the lack of understanding the history or the current tech. Thus the original reason for my first comment.

 

This strikes me odd like people that oppose the SAMS as it is not TI but embrace the F18 when it is not by TI?

 

Why not build a VDP chip set that gives us more VDP and VGA output, so you know we could use FW 80 column or Y.A.P.P. or TELCO 80 or the higher graphics modes available?

Edited October 9, 2013 by RXB

[Tursi]

This strikes me odd like people that oppose the SAMS as it is not TI but embrace the F18 when it is not by TI?

 

 

With all due respect, Rich, I believe it seems odd because you are misunderstanding something. I have seen zero instances of people "opposing" SAMS because it's not TI, only some comments that it is not well enough distributed for some developers to wish to support.

[Omega-TI]

 

This strikes me odd like people that oppose the SAMS as it is not TI but embrace the F18 when it is not by TI?

 

Why not build a VDP chip set that gives us more VDP and VGA output, so you know we could use FW 80 column or Y.A.P.P. or TELCO 80 or the higher graphics modes available?

 

It's not like we're against it, it's just that:

(1) They are almost impossible to find

(2) If you can find one they cost a bloody fortune

(3) Almost nothing exists for them, which makes it hard to justify buying or, let alone getting interested in one.

 

Now as for the F18A... MY monitor say's it's VGA (proof below)

 

 

Sorry for the poor quality, it was taken with my cellphone, but you get the idea.

Oh, one more thing... the F18A is affordable, available and everything works with it, and looks better too, so it's easy to embrace and justify the purchase. Just my opinion.

Edited October 9, 2013 by Kevan

[matthew180]

Why not build a VDP chip set that gives us more VDP and VGA output, so you know we could use FW 80 column or Y.A.P.P. or TELCO 80 or the higher graphics modes available?

Hmm. Let's see:

 

1. The F18A was meant as a replacement for the 9918A, NOT, I repeat, NOT NOT NOT the 9938. The 99/4A console will NEVER, and can NEVER, have 9938 VDP without physical modification to the motherboard, i.e. soldering or cutting some traces. The physical interface does not exist in the 99/4A to support a 9938 via the 9918A socket, thus it is not possible to make a VDP that is *plug-in* compatible, yet guarantee support for the programs you mention, or any other software for that matter.

 

2. The programs you mentioned are simply 80-column programs that use the 9938's expanded memory. Hardly worth the effort of building a 9938 VDP. Where are all the awesome games and graphics-pushing programs that take advantage of the 9938, the extra VRAM, the built in commands, the "awesome" 9995 CPU, etc.?? No reason to go through the effort for 0.0 existing software support.

 

3. "More VDP" you say???? You mean like a built-in 100MHz GPU, horizontal and vertical hardware scroll registers, 64 12-bit programmable color palette, 4 and 8 color sprites and tiles, high speed 32-bit counters, high speed 32-bit random number generators, 30-row mode, tile over sprite priority, true bitmap layer, line-programmable horizontal interrupt, sprite linking, independent sprite size, etc. None of that counts as *MORE* VDP? I guess you were only talking more VRAM to support more "text" pages in VRAM for "text" editors. Lame.

 

4. Ah yes, higher resolutions... The problem is, no one would use them and the 99/4A is not fast enough to handle anything more than is already has. The VDP fits the machine. I could see maybe something like 384x288 max, but that introduces a *lot* of problems because coordinates don't fit in one byte any more, so you blow all existing software compatibility, or make a bunch of "hacks" like the 9938 did which makes it a pain in the ass to use.

 

5. You act like having this stuff available means "everyone" is going to jump on it and start using it. Build it and they will come. Whatever. Believe what you want, but it won't happen. The F18A has been out for over a year now, and I really appreciate all those who have written software to take advantage of some of its features, but the consensus is that compatibility is more important than new features. I have yet to see an F18A-only program.

 

6. Honestly I don't really care about any of the programs you mentioned. I have never seen them run and never had any hardware that could run them. I suspect they require a Geneve (which is not a 99/4A) or some other rare or obscure video card for the PEB. Are any of them games? Do they let me play games? Do 3D graphics? Play music? The LAST thing I want to do on my 99/4A is "edit text"...

 

When I started the F18A I solicited for feedback, so where are your suggestions? What should I have done with the F18A to make it "more VDP"?

[Omega-TI]

When I started the F18A I solicited for feedback...

 

You must have listened well, as this thing is just PERFECT, I would not change a thing! GOOD JOB!

[slinkeey]

The LAST thing I want to do on my 99/4A is "edit text"..

 

 

I am with you here. It is just not an efficient machine for it. My PC is just more efficient. Better laid out keyboard, copy and paste, and etc... You can edit text much faster on the PC and send it to the TI if needed. Sure my PC is probably efficient at many things, but I’d rather have fun with the TI instead of spending time editing some text that I can edit on the PC in much less time.

[RXB]

Hmm. Let's see:

 

1. The F18A was meant as a replacement for the 9918A, NOT, I repeat, NOT NOT NOT the 9938. The 99/4A console will NEVER, and can NEVER, have 9938 VDP without physical modification to the motherboard, i.e. soldering or cutting some traces. The physical interface does not exist in the 99/4A to support a 9938 via the 9918A socket, thus it is not possible to make a VDP that is *plug-in* compatible, yet guarantee support for the programs you mention, or any other software for that matter.

 

2. The programs you mentioned are simply 80-column programs that use the 9938's expanded memory. Hardly worth the effort of building a 9938 VDP. Where are all the awesome games and graphics-pushing programs that take advantage of the 9938, the extra VRAM, the built in commands, the "awesome" 9995 CPU, etc.?? No reason to go through the effort for 0.0 existing software support.

 

3. "More VDP" you say???? You mean like a built-in 100MHz GPU, horizontal and vertical hardware scroll registers, 64 12-bit programmable color palette, 4 and 8 color sprites and tiles, high speed 32-bit counters, high speed 32-bit random number generators, 30-row mode, tile over sprite priority, true bitmap layer, line-programmable horizontal interrupt, sprite linking, independent sprite size, etc. None of that counts as *MORE* VDP? I guess you were only talking more VRAM to support more "text" pages in VRAM for "text" editors. Lame.

 

4. Ah yes, higher resolutions... The problem is, no one would use them and the 99/4A is not fast enough to handle anything more than is already has. The VDP fits the machine. I could see maybe something like 384x288 max, but that introduces a *lot* of problems because coordinates don't fit in one byte any more, so you blow all existing software compatibility, or make a bunch of "hacks" like the 9938 did which makes it a pain in the ass to use.

 

5. You act like having this stuff available means "everyone" is going to jump on it and start using it. Build it and they will come. Whatever. Believe what you want, but it won't happen. The F18A has been out for over a year now, and I really appreciate all those who have written software to take advantage of some of its features, but the consensus is that compatibility is more important than new features. I have yet to see an F18A-only program.

 

6. Honestly I don't really care about any of the programs you mentioned. I have never seen them run and never had any hardware that could run them. I suspect they require a Geneve (which is not a 99/4A) or some other rare or obscure video card for the PEB. Are any of them games? Do they let me play games? Do 3D graphics? Play music? The LAST thing I want to do on my 99/4A is "edit text"...

 

When I started the F18A I solicited for feedback, so where are your suggestions? What should I have done with the F18A to make it "more VDP"?

Ummm the TIM card was a PLUG IN REPLACEMENT for the 9918 with only one ground solder connection using a 9958.

The SOB also gave you a on board Cataloger and fixed the cartridge GROMS so it properly displayed all carts on the system.

 

Other VDP upgrade cards ran from the PBOX or were side plug in.

 

Also Y.A.P.P. for example made TI Artist look like a BASIC graphics. Oh and was insanely fast. (4 times the pixels and 256 colors with Sprites and like the F18 few issues)

XHI did everything you would see in The Missing Link but at 4 times the pixels and twice as fast. (Oh with 256 colors to boot and 32 sprites in XB)

X80 gave 80 columns to XB for text so not much use today I agree.

Funnel Web 80 displayed a really PC like screen with all the options of FW so despite the TEXT Editor did look great with more than 16 colors.

 

Just because you never see something you can poo poo it?

 

You started the F18 before I came back to the TI community so that is the reason I did have any input. Making suggestions is not a crime is it?

 

Models for these cards are many versions like AVPC or EVPC or TIM are some I know were popular but others were also built and sold.

 

I realize you did hard work on the F18 but I was just making suggestions to update the TI a little as it is not like it is the first time that has been suggested is it?

[matthew180]

Ummm the TIM card was a PLUG IN REPLACEMENT for the 9918 with only one ground solder connection using a 9958.

Not true. I just read the install document published by OPA for the TIM, pg3:

 

"... this wire is being attached to A13 of the system address bus, used by the V9958 in decoding the new memory-mapped I/O ports."

 

Just like I said, you will never add a 9938 or 9958 to the 99/4A without soldering on the motherboard. The F18A was a pin-compatible plug-in replacement without ANY exceptions. The TIM also has problems, just as I mentioned, with existing software because the 9938 and 9958 are only "software" compatible with the 9918A. The install document goes through great lengths to explain this, going so far as to even suggest sector editing software available on disk. Poorly written software for the 9918A will not work on the 9938 and 9958 VDPs. The F18A specifically has a lock-out to prevent this poorly written software from failing. I also had other systems besides the 99/4A to think about. Try to plug the TIM into a ColecoVision, MSX1, Baby-PacMan coin-op, etc.

 

I'm not bashing the TIM, I'm sure it was a cool upgrade. I'm simply pointing out that it is NOT the same kind of upgrade as the F18A. I could have easily made the F18A give you access to 648x480 graphics and all kinds of incompatible stuff, but it would not work with ANY existing software, so what good would that be? If you think people would come out of the wood-work and start producing brand new software just because you have more pixels then I have a bridge to sell you.

 

The SOB also gave you a on board Cataloger and fixed the cartridge GROMS so it properly displayed all carts on the system.

The SOB also needs a ground wire soldered to the motherboard. Now you are up to two solder mods *AND* it replaces two of the original system GROMs. Of course with completely rewritten system GROMs you can do anything you want. Not very useful for software developers though, since 99.9% of the people with 99/4A consoles won't have an SOB or TIM.

 

Just because you never see something you can poo poo it?

I'm not poo pooing it, but I can't run any of it either since it requires hardware that hardly exists and is not available. Just because you can't run your 9938/58 only software on the F18A means you can poo poo it?

 

You started the F18 before I came back to the TI community so that is the reason I did have any input. Making suggestions is not a crime is it?

No, but I think you were around since it took me two years from inception to delivery, and I fixed your GRAM boards sometime in the middle of all that. But your comment was "why not make more VDP", which you still have not told me what exactly that means, or what "more VDP" would have entailed.

 

I realize you did hard work on the F18 but I was just making suggestions to update the TI a little as it is not like it is the first time that has been suggested is it?

So you don't consider the F18A an update? What is it missing? You still did not tell me what I should have done or added. You also did not tell me why the output generated by the F18A is worse than the 9938/58.

[+Ksarul]

Matthew, I just saw confirmation that the F18A works fine in a Powertran Cortex (I still haven't had time to put one of mine into my Cortex, but Jim Hearne installed one in his). It needs the long pins. I'll probably put one into a Pyuuta Mk II next (replacing a TMS9118), since I have two of the computers to experiment with and can thus keep one unmodified.

Edited October 10, 2013 by Ksarul

[matthew180]

@Ksarul: Cool! I'm always happy to hear that the F18A works as expected in another system. Do you know if Jim has any photos posted somewhere?

 

I looked at the 9118 pinout and the F18A should work fine. I didn't even know such a chip existed until someone emailed me about it long ago (was it you?)

 

The main changes in the 9118 are improved DRAM addressing, which the F18A does not use. The 9128/29 do move the CPUCLK to the GROMCLK output, which may be a problem if your original system uses a 9128/29 and requires the CPUCLK on pin 37. However, I could fix that with a firmware update and use one of the two unused jumpers. Please let me know how it goes.

 

The 9118/9128/9129 VDP's are the succesors of the 9918A/9928A/9929A VDPs.
These new chips have improved memory addressing circutry which allows the
interface of either 8 TMS4116 (2k) or 2 TMS4416 (8K) dynamic RAMs.

                                           VDP PROCESSOR TYPE
        _________
 RAS  =|1   U  40|= XTAL1          9118    9918A    9128/29 9928A/29A F18A
 CAS  =|2      39|= XTAL2          ======= ======== ======= ========= ====
 AD7  =|3      38|= .............. CPUCLK  CPUCLK   R-Y     R-Y       NC (USR4 on) or CPUCLK (USR4 off)
 AD6  =|4      37|= .............. NC      GROMCLK  CPUCLK  GROMCLK   GROMCLK
 AD5  =|5      36|= .............. COMVID  COMVID   Y       Y         NC
 AD4  =|6      35|= .............. EXTVDP  EXTVDP   B-Y     B-Y       NC
 AD3  =|7      34|= RESET/SYNC
 AD2  =|8      33|= vCC
 AD1  =|9      32|= RD0
 AD0  =|10     31|= RD1
 R/W  =|11     30|= RD2
 vSS  =|12     29|= RD3
 MODE =|13     28|= RD4
 CSW  =|14     27|= RD5
 CSR  =|15     26|= RD6
 INT  =|16     25|= RD7
 CD7  =|17     24|= CD0
 CD6  =|18     23|= CD1
 CD5  =|19     22|= CD2
 CD4  =|20     21|= CD3
       `---------'

Edited October 10, 2013 by matthew180

[Asmusr]

5. You act like having this stuff available means "everyone" is going to jump on it and start using it. Build it and they will come. Whatever. Believe what you want, but it won't happen. The F18A has been out for over a year now, and I really appreciate all those who have written software to take advantage of some of its features, but the consensus is that compatibility is more important than new features. I have yet to see an F18A-only program.

 

 

For my part I haven't made any decision that all software must be backwards compatible. It has more to do with the lack of emulation and to some degree documentation, which means it would take far longer time to develop a game that took advantage of all the new features. Having said that, there is something exciting about knowing that the software you develop today would also have worked 30 years ago. Regarding SAMS, if we had it in the nanoPEB I would definitely have used it for Scramble to avoid the level changing delays.

[matthew180]

... It has more to do with the lack of emulation and to some degree documentation, ...

Very true. Without emulation support it is definitely harder to develop code. Testing the F18A was definitely a pain in the ass. Especially once you get used to developing under emulation. I'm guilty of not providing an F18A emulator, as well as the lack of documentation. It is hard to justify the time though, since the demand is very low and the effort required very high.

 

... there is something exciting about knowing that the software you develop today would also have worked 30 years ago. ...

I agree with that too, which is why the primary design goals of the F18A were targeted at simply replacing the 9918A and not trying to produce something such as the 9938/58. The enhancements were simply because I had room to put them in. It is a slippery slope since with the cheap hardware available today it is possible fix all the problems with the 99/4A, give it massive expansion, mega storage, etc. But then it would not be a 99/4A any more and no one else would have one, and it might as well just be an Amiga or some other system that came afterwards.

 

Some of the F18A features are over board, but others I think the 9918A would have included (single page hardware scroll registers for example) if the engineers were afforded a little more time and transistors.

 

Regarding SAMS, if we had it in the nanoPEB I would definitely have used it for Scramble to avoid the level changing delays.

What about availability? The CF7/nanoPEB are still not easily obtained, so your software would rely on people having that hardware. I could make the argument that you could fix the level delays with the F18A GPU, or use the hardware scroll registers to afford yourself more time to decode levels on the 9900. Either solution (SAMS or F18A features) requires hardware that was not available with the original 99/4A system.

[Asmusr]

What about availability? The CF7/nanoPEB are still not easily obtained, so your software would rely on people having that hardware. I could make the argument that you could fix the level delays with the F18A GPU, or use the hardware scroll registers to afford yourself more time to decode levels on the 9900. Either solution (SAMS or F18A features) requires hardware that was not available with the original 99/4A system.

 

Well, I could use SAMS if it was available, and fall back to delays. I don't think this could be fixed with the GPU because we're talking about CPU RAM -> CPU RAM or CPU RAM -> VDP RAM transfers. Even with the scroll registers I would (probably) still need to keep the maps in CPU RAM, but they could be packed differently, in columns instead of rows, like in the original, so you would only need to unpack on column at a time. It would be fun to make a version of Scramble that used all the F18A features, but I would have to sit in front of the TI moving CF cards around instead of in my sofa with my laptop.

[matthew180]

Well, I could use SAMS if it was available, and fall back to delays. I don't think this could be fixed with the GPU because we're talking about CPU RAM -> CPU RAM or CPU RAM -> VDP RAM transfers.

I don't know the details, but maybe pass the compressed data to the VRAM and let the GPU unpack it. The CPU -> VRAM has to happen no matter what. I don't know where the main speed problems are though. When you say "decompress" I think of a time consuming process which the GPU would be really good at dealing with. The GPU also has 2K of private RAM to work with and can access VRAM as 16-bit data. Something to think about anyway. Also, Classic99 implements the GPU and supports it in the debugger.

 

Even with the scroll registers I would (probably) still need to keep the maps in CPU RAM, but they could be packed differently, in columns instead of rows, like in the original, so you would only need to unpack on column at a time.

Something else to keep in mind is that the F18A has a programmable VRAM "address increment" register. So you can write to VRAM and have the address pointer increment by any value from -128/+127 (signed byte). So you can write name table columns to the VRAM just as fast as rows (at least for a single column) by setting VR48 to 32.

 

It would be fun to make a version of Scramble that used all the F18A features, but I would have to sit in front of the TI moving CF cards around instead of in my sofa with my laptop.

That is true. Unless you are a perfect programmer and can get it right the first time. :-) Classic99 can help a little with the GPU part if you use the GPU. I can't do much for the CFcard swapping problem though, but it is a *big* pain for me too so I'm motivated to resolve the issue.

[Omega-TI]

 

... but I would have to sit in front of the TI moving CF cards around instead of in my sofa with my laptop.

 

With CFHDXS1 and that Bluetooth to RS/232 device... the day's of CF swapping may become a thing of the past.