Matt_B

Needs more woodgrain.

30-50fps at 720p on the lowest setting seems to be the general consensus of what the Vega 3 is good for with Fortnite: https://www.userbenchmark.com/PCGame/FPS-Estimates-Fortnite/3954/511366.0.0.0.0 A small number of people managed to get it up to 60, but I wouldn't be surprised if that was with the aid of a combination of overclocks, faster RAM, running at 540p, config hacks or just leaving the camera pointing skyward. 4K is going to be in single figures at any settings. Those numbers are mostly based on the Athlon 200GE and the Ryzen 3 2200U rather than the 1606G but it's the same GPU and dual core CPUs of similar clocks, so I doubt there will be much difference.

It would be quite hilarious if the VCS was to get snapped up as an off the shelf solution for the digital signage market. That's already happened with the Shield TV to a certain extent. It's cheap, very easy to set up, and you can leave it running practically indefinitely. I'm not sure how well the VCS will score on those things though.

Quite obviously, it's way more complicated than that. For starters, you couldn't even put a PC made out of existing parts in the first case they showed. The ports at the back didn't line up, there was no way you keep a 35W chip cool in it and it was always going to need a custom motherboard. A bunch of us said so at the time on the other, now locked, topic. So, it came as no surprise when after a year of nothing much happening, they changed the case, the ports layout and the chip. Basically, nothing that they presented at the crowdfunding has survived. That's why you need a prototype, and why Atari are a bunch of amateurs. AMD will make sure the chip works with Windows and provide a reference board. That's pretty much it. By all accounts Intel are far more supportive when it comes to making custom PCs.

Sure, crowdfunding is about raising money. However, there's still the expectation that it's a product that's going to be delivered and that the people behind it have the competence to do so. I'm pretty sure that a device like the Spectrum Next couldn't exist in the traditional sense either - it was not so much aiming for a niche, but a crack in the wall with only around 3000 backers on Kickstarter - but that didn't stop them from building a working prototype before they went to crowdfunding. If you're working with off-the-shelf chips it should only cost a few thousand dollars too so you could go through several iterations of this before settling upon a design that's right for the job, at least if you know what you're doing. Atari obviously didn't know what they're doing and they're also cheapskates into the bargain. Surely I don't need to point out he difference between the barely working prototype of the ST and a lump of plastic? The former nowadays might just have been good enough to satisfy Kickstarter's requirements but the latter was never going to. Atari went with Indiegogo precisely because they're the Wild West of crowdfunding and allow you to take money for pretty much anything at the glint-in-the-eye stage. Um... what's the best game on AntStream? 🤣🤣🤣

Sure, product development can be messy, but there's no requirement to launch a crowdfunding campaign when all you've got to show is a few lumps of plastic and some mocked-up footage. Rather, most people serious about making hardware would construct a working prototype first - using their own money and behind closed doors like you say - and the fact that Atari didn't do this is what shows them to be a bunch of chancers. As such, it's entirely down to their mismanagement that we've had such a good view of its underbelly and even then one of the most common complaints is that they rarely put out substantive updates about the true status of the project which is why it's still a matter of guesswork as to whether they'll be able to ship or not in a few weeks.

That's what you get for repeatedly delaying something that was originally envisioned to come out in 2018. Anyway, not to worry. They'll probably just delay it again and avoid the clash that way. 😀

It was always confirmed to be Linux-based but not an open system. Rather, if you wanted to install your own software, you always had to go into 'sandbox' mode and bring your own OS. It was certainly not explained how this was going to be implemented - by adding additional storage and having the option to boot from it - until Rob Wyatt joined the project, and a prototype not consisting of a solid lump of plastic was being worked on. I'm not sure if his explanation would still hold now that he's left it even if Atari haven't said anything else to contradict it. Also, if Atari have have ever shown it booting into Linux under Sandbox mode, it's passed me by.

What have you seen? So far as I can tell everything they've shown us has been mock-ups or a PC running Windows.

Ah, there's nothing I've love more for Atari to get the thing out and to be able to buy them internationally and flip for the Australian retail price at a huge profit. Alas, I'd suspect that there are probably going to be issues on both the supply and demand sides of that. I.e. Yet another unannounced delay seems more than likely, and there probably aren't too many people wanting to pay $100 more than the price of a Series X for one.

Yeah, it's not like they haven't just ghosted past deadlines before. I'd think that anyone likely to be disappointed by an after-the-fact announcement of a delay has probably already lost it with Atari by now anyway, so there's not really any incentive for them to change.

The Pi 3 and 4 have built-in WiFi and Bluetooth so no dongles are needed and you're not going to have the same sort of driver issues. That pretty much negates the need to plug anything into the Ethernet and USB ports once you have it set up. I'm surprised you'd have problems running Stella on a Pi 2, as it works just fine on mine. I've only ever used the default install on RetroPie though, so there may be config settings that you have to change to get the performance up. The Pi 2 does have some performance issues with more demanding 16-bit emulators and needs frame skipping to handle anything beyond that but it's always done a pretty good job with all the 8-bit systems I've thrown at it. I've also had no issues with Amiga and ST emulation on RetroPie either, although I do keep my games collection as a set of floppy images rather than going for emulated hard-drive installs. Disk changes need some menu navigation, but that's about it. You might not have had many problems with Linux on x86 PCs of late, but I can assure that they're not entirely free of problems. Support tends to be good with most desktop chipsets, and laptops from the big brands, but custom-built small production run PCs can still be an utter pain. For instance, I had a good go at trying to get it running on my GPD Win and... yeah... not only did the ACPI not pick up but the screen was rotated round 90 degrees, the WiFi didn't work and the only thing the USB ports were good for was charging. Apparently the Win 2 works much better though, so there's probably an element of luck of the draw with these things.

I was mainly talking about why the Pi managed to find a niche in an otherwise Mac/PC dominated computer market, and it's mostly about it being very much cheaper than the existing options. Being physically small and having low power consumption helps too, but it didn't just dribble out after a succession of missed deadlines into a market crowded with similar products flaunting a vintage case and logo as its only selling points. I'm genuinely curious as to what are you trying to emulate that the Pi is too slow for? Except for the original model and its derivatives like the Zero, they're pretty decent and easily make a better emulation box than other commonly used devices such as an Ouya, a hacked Xbox/Vita/Wii or almost all the handhelds in the sub $200 range. There are PS1, Saturn, N64 and Dreamcast emulators that can run games at full speed on a Pi 4 and ready made distros (RetroPie, Lakka, etc.) with RetroArch and a bunch of cores that you can just slap on an SD card and be up and running with in minutes. Compiling your own isn't necessary unless there are things in the source that you want to change. I don't think you should necessarily count on the VCS to do any better at power management unless you put Windows on it. Linux is not totally guaranteed to be able to access ACPI features even if the hardware has them and that might well include the VCS because it's based on a chipset that's still pretty new. A Pi 4 at idle consumes power comparable to a PC in sleep mode anyway so always-on is an option, or you can pay $20 extra for a remote control with shutdown and power-off features if it's still a total deal breaker for you. Do you really think the Pi requires too much cabling? The ones I use as media/emulation boxes have precisely two cables both on the same side; one for the power and one for the HDMI, with everything else over WiFi and Bluetooth. I'm pretty sure the VCS won't do any better. It's nice to have things like USB, Ethernet and the audio jack for tinkering purposes but it's not like you're going to use them when the device is set up. The Pi isn't the be all and end all of emulation boxes by any means but it does set the bar above which you can legitimately ask what you're paying extra for. I don't see the VCS bringing too much extra to the party, and probably nothing at all unless you're going to put an SSD in it and install Windows, in which case building your own won't require much more effort and would give you better results for the money. If Atari knew what they were doing, I don't think they'd be making the VCS at all, and a good look through the history of the device ought to confirm that they don't know what they're doing. As it stands, they know that they can sell 11,000 crowdfunded units plus whatever preorders they had, but I honestly don't think they should be counting on much more than that. Still, as others have said, I won't complain about overproduction if it drives the price of second hand units through the floor as that's the only way I can see myself picking one up.

I don't think Atari had to compete against the big three, although that's not stopped them making numerous such comparisons in their marketing over the years inviting others to do so in kind. Rather, I'd think they'd have to look for a niche that would allow them to co-exist with the big three and I don't think they've got one. Certainly, when you look at the sort of niche devices that have been at least moderately successful of late, they're things like the Linux/Android/Windows gaming handhelds like the Dingoo, Nvidia Shield and GPD Win, the sub-$100 plug-n-play retro consoles in which I'd include the Flashbacks, the even cheaper single-board PCs like the Raspberry Pi, FPGA devices like Spectrum Next and MiSTer, and cartridge-compatible retro-consoles like the RetroN series. Success isn't necessarily always measured by sales, but I'd think that you have to be offering something substantial that people can't just already get from a PC or games console at around the same price. Not repeatedly delaying it while also hiding the details of its development behind a veil of secrecy probably helps too. You're also giving Atari more credit than they deserve if you think they're going to be engaged in actual sales forecasting. I'd think that the 11,000 or so crowdfunded orders plus whatever extra they can glean from listing it on various online stores - which presumably won't be that many given the bad publicity, lack of unique features and high price - will be pretty much the extent of their manufacturing run.

I wouldn't knock the Series S - or the One S for that matter - for casual/retro gaming. Although it's obviously AAA focused, there are loads of cheap and simple casual games, local multiplayer options, retro compilations and media/streaming apps in the Microsoft Store. The Switch maybe gets a better selection, but not by that much; you've really got to want Nintendo's own games and/or the ability to pick it up and take it with you before it starts to look a clear winner. For anyone who thinks that Flashback Classics, Antstream and Plex are selling points for the VCS, the Series S will do them all too. I do most of my retro/casual gaming on the PC anyway. The Switch is for AAA on the go. 😀

Obviously you're not old, but you are pretty much the definition of a brand loyalist if you'd think of getting something because it's got the right badge on it rather than whether what's inside is any good or not. Believe it or not, but I didn't buy every single piece of Atari (or Sega) hardware despite growing up during their heyday. Rather, machines like the original VCS, the 8-bit home computers and also the ST (because 'real Atari' or not, the Tramiels at least knew something about computers) were worth having because they were impressive hardware at launch; it's not like you could go to their competitors for something much cheaper and with better software and games like you can nowadays. Some of the other Atari hardware wasn't so great though, and that tended to sell very poorly in comparison, so I obviously wasn't the only one who saw them that way. I've no great dislike for the Lynx either, other than the ferocious rate at which it consumes batteries. Epyx made great games throughout the 80s, so what's not to like about them designing a handheld? If there was someone with that kind of pedigree behind the design of the (new) VCS, or it was offering similarly good hardware for the money, I'd probably be a bit more enthusiastic about it. However, it's the brainchild of a guy whose best bit of tech on his resume is a Minecraft wristband, who had to sue Atari to get paid, and it's low-end off-the-shelf technology at an overinflated price. Sure, it's their brand and they can do what they want with it, but we don't have to like it. I think you've got me wrong there. I was responding to Bill's post and trying to contrive an, albeit unlikely, scenario in which the VCS ends up selling more and that's about the only way I can see it happen. As he concluded, if both machines hit the market at around the same time it's hard not to see the Amico selling more seeing as it's the one that's actually going to get some exclusive games.

An easy to envisage scenario is that the Amico backers take stock of the market they'll be launching into after Microsoft dropped that bombshell, and decide to pull the plug on the hardware. The preorders get cancelled, the games get ported over to existing systems, and not too much money goes down the drain. Atari could then score a Pyrrhic victory by eventually dribbling out enough units to fulfill the crowdfunding. They can then cancel the other preorders and who cares about porting the games because they never made any. That sounds plausible enough to me at least, although I'm sure reality can come up with something even more ridiculous. Yeah, they're like one of those Chinese car companies that bought a defunct marque from twenty years ago so they can sell to befuddled old brand loyalists.

Price aside, I'd think that the Series S holds another serious advantage in that Microsoft are almost certainly going to hit their street date with it.

Atari have been very cagey about the OS on the VCS so far. It's certainly going to be Linux based, but whether they'll give you root access to install your own applications or get at the bootloader to put a custom OS on there remains an unanswered question. All that's really guaranteed so far (at much as anything ever is with this project) is that you'll be able to boot it into another OS if you put an additional SSD in there. Linux is far from guaranteed to be an easy install either and it usually takes a bit of of work to get all the drivers ironed out on new hardware, and I wouldn't be expecting Atari to do more than the bare minimum to get their own OS working. Windows, on the other hand, is going to work because AMD are obliged to support it. Also, what model of Pi are you using, if you don't mind? The original was certainly very slow and of only marginal performance when it came to emulation or graphically intensive operations, but since they made the move to four cores they've been much snappier. The Pi 4 is generally considered about on a par with a Core 2 CPU from about a decade ago, so unless your idea of retro is newer than the Dreamcast and N64, it ought to be up to the task. Rather than building emulators for the Pi individually, I'd suggest installing the RetroPie or Lakka distros that come with their own front ends, RetroArch and a bunch of cores for all the common systems; they'll all have been tested so all you'd have to do is supply BIOS ROMs and the games themselves to get up and running. Unless you're emulating something very obscure you shouldn't have too much trouble. RetroPie also has Kodi with the more recent builds so you don't have to reboot to use your Pi as a media box either. Anyway, yeah, wait and see. There's usually just a lot of waiting with Atari these days and not much seeing.

I'm not saying that the Pi is going to be a useful system for everyone. It's a very basic computer after all, so things like power control are left to users to implement if they really want it. A switch on the power supply does the job for me but there are fancier remote control solutions if that's what you want. You've certainly got a lot of scope for upgrading it before it starts to creep into the VCS price range though. The point is rather that it can manage all the important features that the VCS is confirmed to ship with, i.e. Plex, AntStream and emulation of older Atari systems, at a fraction of the price. It'll also cover some other low hanging fruit that might later be added if Atari can ever get around to it, such as emulation of non-Atari systems up to the Playstation era, Netflix, Kodi and Steam Link. Or you can boot it in ChromeOS or Linux and use it as a desktop PC. Irrespective of the power of the underlying hardware, the Pi already does more for less than the VCS promises to do out of the box. Rather to get more out of the VCS you are looking at having to mod it by adding an SSD and installing Windows, at which point it's worth considering what your other options are in the way of PC hardware. No, you're not going to get a killer gaming PC for $390, but you could at least build the bones of a system that can later be turned into one and will still outperform the VCS for now. That much ought to get you the aforementioned 3200G, a B450 motherboard, 16GB or RAM and a 240GB SSD. You won't get the case of your dreams and might have to settle for a couple of no-name (PowerA aren't exactly a high-end brand either) controllers but you've at least got the upgrade potential to add a decent graphics card and a better CPU at some point down the line to keep up with trends in PC gaming. The VCS will forever be gimped by the 1606G being soldered to the board though. I won't deny that there's something cool about the case. However, once you consider how little the current owners of the brand have to do with previous Atari hardware and look into the three year rolling car-crash of its development so far, that aspect rather starts to lose its sheen too.

Any PC can play some games though. The point is to be getting something that's worth what you're paying for it. The GT 710 is a low-budget card. It's all right paying a sub-$40 price for that level of performance when you can use it to give a new lease of life to an old potato PC. Similarly, the Raspberry Pi is an entire computer for around $40. I've bought four of them over the years because I like tinkering. Atari are asking $390 for the VCS with a couple of controllers. For that money you ought to be getting something capable of playing modern games at decent settings. You can, after all, buy a PS4 for less.

I'd hardly say it's much more powerful than first mooted. Both the A10-9630P and the Ryzen 1606G are off-the-shelf chips, so have been widely benchmarked. On the plus side, there's about a 50% improvement in CPU performance: https://www.cpubenchmark.net/compare/AMD-A10-9630P-vs-AMD-Ryzen-Embedded-R1606G/2857vs3659 However, the GPU is actually slightly weaker than the one on the A10: https://www.videocardbenchmark.net/compare/Radeon-Vega-3-vs-Radeon-R5-A10-9630P-4C+6G/3926vs3785 On the whole, I get the impression that the Ryzen 1606G would be more at home in a smart fridge than a games PC.

You didn't see a Ms Defender or Ms Robotron in the arcades back then for a reason. 😄 Still, I'd agree that it's mostly about the casual/hardcore divide these days and flagging games as particularly for boys or girls is likely to get you severe pushback from some quarters.

Does anyone think they're actually going to make the November release date? Let's face it. They'd need to have made at least 10,000 of them, gone through QA, boxed them up, and packed the boxes into containers that are now on board ships, without any of it having been photographed or videoed and shared with us. There's also the mystery about the '96' units that have been made, in that not even one of them has been seen plugged in and turned on yet. As such, the expectation is that they don't work. It might just be that they're missing a finished OS or have some easily ironed out glitches that they're rather not share, but it's hard to shake off the suspicion that there are more deep-seated issues still waiting to be resolved with them. That's my main issue with the Plex announcement. If the project was clearly on track it'd just be an extra feather in their cap - that people could either take or leave depending upon interest - but against a backdrop of continuing delays, lawsuits and an unusual level of secrecy for a crowdfunded venture, it just serves to highlight the things that they aren't telling us about. So, my recommendations to go with a Raspberry Pi or Windows PC over the VCS are more about pointing people in the direction of things that they can buy now, without waiting for a pack of unreliable shysters to get their act together. I'd love nothing better than to be proven wrong, and for them to hit that deadline. However, there are still more red flags than a communist party shindig about it.

I'll give you some serious answers. Atari seem to think Antstream, Plex and emulation of 2600 and Atari arcade games are their biggest selling points for the VCS, because they've not announced much else. You can do all of those on the Raspberry Pi right now; they're not hypothetical things that it might get in the future like Windows on ARM. You can run full versions of Linux and ChromeOS on the Raspberry Pi too. You are currently limited to (free) Windows IOT, but when you factor in the extra costs of an additional SSD and licence, the VCS isn't exactly going to end up much like a cheap option for a Windows device. You can buy dedicated Windows nettops/HTPCs for around the $200 mark though, if that's really what you want. In terms of emulation, the Raspberry Pi is generally good enough for anything up to the PlayStation, including all the consoles and computers Atari ever made. Dreamcast, Saturn and N64 emulation is a bit spotty but works well enough for a lot of games. That only leaves the PSP, DS, PS2, GameCube and Wii in terms of significant consoles that can be emulated well on a high-end PC but not the Pi. You will need an additional SSD to attempt to emulate any of those machines on the VCS though and the performance will not be as good as a decent Windows desktop, or even a mid-to-high laptop, because it's only got a low-power dual-core APU that's mainly designed for embedded applications. Still, I get the advantages of being able to run Windows on an HTPC. I already use one that I first built in 2014 - to a fairly similar spec to the VCS although with much more storage - and it's still capable of doing a few things that the Raspberry Pi can't. However it's a much shorter list now than it was back then, and it wouldn't be much good for games any more if I hadn't been able to upgrade the graphics card. If flexibility is truly what you value, you're better off building your own.