11 ms·
HDMI ISA graphics card for vintage PCs by improving the Graphics Gremlin
- prmoustache 3y ago"the brown colour is displayed incorrectly as dark yellow" I understand what it means in an RGB context but it is the first time I seesomeone mentionning dark yellow as a color.
- marceldegraaf 3y agoThere's a lot to say about the color brown :-) https://www.youtube.com/watch?v=wh4aWZRtTwU https://www.youtube.com/watch?v=wh4aWZRtTwU
- KingLancelot 3y ago[dead]
- boffinAudio 3y agoI imagine a PC with a Graphics Gremlin and either a Snark Barker or Snood Bloober for audio, is going to be quite the hipster toy to have, one of these days .. all this great old, retro, PC hardware designs!
- jamesfmilne 3y agoSounds well Jackson.
- LeoPanthera 3y agoI see that you are a self facilitating media node.
- cmpxchg8b 3y agoI got a Wasp T12 Speechtool - it's well weapon
- waddup_ 3y agoI refuse to believe Snark Barker and Snood Bloober are real device names, what a funny time in computing that was.
- rsynnott 3y agoThey're modern clones of the SoundBlaster 1 card (a real late-80s sound card).
- boffinAudio 3y ago"Sound Blaster" was the original funny name - these are both derivations in a modern style, given to clones of that ol' sound card .. and yeah, they are hilarious names, and I hope the trend sticks, personally .. the world tires of X3000-style naming conventions, I'd wager, and for fun things like sound cards and video cards, a bit of naming whimsy would be welcome.
- LeonM 3y agoKids nowadays will never understand how cool it was to have a Sound Blaster, an Athlon Thunderbird and Voodoo graphics. I might be nostalgic, but to me it sounds much cooler than having a 11700k with a 4090 nowadays
- zare_st 3y agoThere was a time before Pentium when hardware was not named with trade name but by standard it implements. For instance you had your 486 with VGA, were it Cyrix CPU with Trident graphics card or Intel CPU with S3 graphics card didn't matter much. For consumers only sound expansions broke this mold because they came with cool names such as Sound Blaster, Gravis Ultrasound, etc. They carried that over from audio market where equipment had "cool" names for some time already (since 70s at least). Interestingly, Intel made "Pentium" trademark to distance and distinguish themselves from other x86 CPU vendors, but Creative failed to protect Sound Blaster trademark in a way that in mid90s everyone called every OPL3 card "a sound blaster". In that age real Sound Blasters had quality synthesizer, big wavetable with a quality soundfont and a DSP for effects on top of it, a feature which became less and less relevant as 90s went by. In mid 90s games were already moving to MIDI+CD Audio combination and by late 90s MIDI based soundtracks were largely gone. When you were playing Duke for the first time the type of "sound blaster" you had and hardware around it mattered. A clone would render worse MIDI than real Creative hardware and then came the external hardware such as Roland SC series, etc. The differences were real. When you were playing Half Life or UT for the first time just a few years after, there was no difference. So Creative never managed to actually stamp "Sound Blaster" with something particular. It always remained a synonym for the entire standard and something even cheap hardware was described by. "Sound Blaster Audigy" was mostly reffered to as Audigy or Creative Audigy and it didn't have an ubiquitous meaning because it was clearly separated from all the junk audio cards of the day.
- LeonM 3y agoI've worked on something similar while in college. HDMI is actually a fairly simple protocol, based on VGA. It still has scanlines and such, but the signals are digital instead of analog. So instead of a RAMDAC, you have the HDMI encoder chip. You typically need control some i2c interface to set the HDMI encoder chip in the correct mode, but that's about it.
- unixhero 3y agoI remember i2c support was added to the Linux kernel. I never understood what it is. Next in line is BPF, which is also unclear.
- irdc 3y agoI2C is a simple serial communications protocol typically used for things like configuring on-board devices and reading sensors. BPF is anything but simple.
- elric 3y agoIf you're talking about (e)BPF, the (extended) Berkeley Packet Filter, the easiest way to think about it is like a tiny virtual machine running inside the kernel, which can execute "simple" commands that would otherwise be very slow or very complex from within userspace. The traditional example would be counting the number of packets being sent out by a network interface. But it turns out that eBPF is massively more general purpose than that, allowing people to develop all kinds of monitoring applications.
- spicyjpeg 3y agoI2C is basically "USB for things where USB would be overkill". It's a standardized protocol [1] that allows for one host to communicate with up to 128 different devices on a shared serial bus consisting of two wires. The specification does not define any discovery mechanism or higher-level functionality on top of basic packet sending and receiving, so I2C is meant to be used in cases where the host already knows what is connected and has the appropriate drivers set up (hence why embedded Linux devices typically use a "device tree" or similar configuration file to tell the kernel about their I2C hardware). Many of the chips used in modern electronics are either controlled entirely via I2C or use it as a configuration port in addition to a separate high-bandwidth interface for data; the accelerometer, power management chip, audio DAC, display and cameras in your phone all have I2C interfaces. Another use case for I2C is device identification: since I2C-interfaced ROMs are so cheap, it's common to embed them in various types of peripherals and have the host read them out to retrieve information about the peripheral. This is how your PC gets to know which resolutions are supported by your monitor [2] (VGA, DVI and HDMI all have dedicated I2C pins for this purpose) and which type of RAM you have installed [3], for instance. [1] https://en.wikipedia.org/wiki/I%C2%B2C https://en.wikipedia.org/wiki/I%C2%B2C [2] https://en.wikipedia.org/wiki/Extended_Display_Identification_Data https://en.wikipedia.org/wiki/Extended_Display_Identificatio... [3] https://en.wikipedia.org/wiki/Serial_presence_detect https://en.wikipedia.org/wiki/Serial_presence_detect
- haunter 3y agoAre there any recommended made to oder PCB services if you just want a card? Googling around seems to be a lot of options all over the world but maybe some HNers have experience with some of them.
- mikecoles 3y agoJLCPCB has worked for me. There are others, but I haven't had the need to look elsewhere. OSHPark is another I recall. https://jlcpcb.com/ https://jlcpcb.com/ https://oshpark.com/ https://oshpark.com/
- geerlingguy 3y agoAnd PCBWay.
- mmastrac 3y agoJLCPCB is great as long as your components don't get too small. They made a few boards for me with a tiny component that just didn't work. I think they vastly overestimate their minimum trace size.
- jan139748 3y agoRelated development for those interested in using old hardware with new displays. There is this ISA HDMI addon card in development that will add HDMI port to your existing old display card: https://www.vogons.org/viewtopic.php?t=92512 https://www.vogons.org/viewtopic.php?t=92512
- gwbas1c 3y ago> The frequencies and connectors used by CGA and MDA are no longer supported by modern monitors hence it is difficult for older PCs of the 1980s era to have modern displays connected to them without external adapters. Some CGA ran over composite, and there's plenty of modern small TVs with a composite input. It's perfectly fine to use a TV as a computer monitor. (I do!) > This analog-to-digital conversion will also lead to an inevitable loss in video quality. Oh, now we're splitting hairs! This is super-low resolution, super-low colorspace. CGA was (at most) 16 discrete colors. In many situations it was 4 colors with 2 pallets to chose from. The CGA port was also digital, so I don't understand where the "loss in video quality" argument comes from. IMO: I don't "get" this. You're no longer running "vintage" hardware; yet a lot of vintage hardware has limited lifespan and may become unrepairable if/when there's degradation inside the chips themselves. If someone is going to go through all this trouble, it makes a lot more sense to emulate the whole computer.
- treve 3y agoCount me in as someone who likes vintage hardware but modern displays. In my case it's old game consoles. The drop in video quality with composite is real. This has less to do with the resolution, but more with the fact that hardware that upscales this to an HD or 4K panel needs to make an educated guess where pixels start and end, and gets it wrong. It looks quite ugly practically and switching to something with crisp pixels is usually very worth it. For old game consoles it's often enough to switch to RGB or Component and you don't have to go full digital. Composite (and RF) are quite bad. This is not an audiophile type of distinction, it's very visible and obvious to almost anyone.
- gwbas1c 3y ago> The drop in video quality with composite is real. Remember, this is CGA. Some games specifically take advantage of composite: https://en.wikipedia.org/wiki/Composite_artifact_colors https://en.wikipedia.org/wiki/Composite_artifact_colors Likewise, remember that the monitor connector is digital. If you build an HDMI (or DVI, DisplayPort, whatever) converter, you're starting with a digital signal, not an analog one: https://en.wikipedia.org/wiki/Color_Graphics_Adapter#Specifications https://en.wikipedia.org/wiki/Color_Graphics_Adapter#Specifi...
- jwells89 3y agoI’d love to see something like this for late 90s through mid 00s computers, because many of those machines can still find modern uses but are awkward to use with modern high-resolution displays, even if they can sometimes be adapted digitally (DVI-equipped machines). Of course you can always grab an old monitor to pair with said machine, but that comes with heavy picture quality concessions in the case of LCDs, and good CRT monitors are becoming rare and expensive. So for example it’d be super cool to be able to drop a new GPU into a PowerMac G4 tower and allow it to drive a modern 2560x1440 display under both OS 9 and OS X.
- jandrese 3y agoJust be aware that old machines can bog way down when you start asking them to push millions of pixels. Their old graphics subsystems were never designed to handle that much data, bus widths are a major bottleneck.
- jwells89 3y agoYeah that’s a good point. There’s definitely some machines in that bracket that are capable since with the right GPUs they could drive similar contemporary displays though (G5 tower with circa 2005 2560x1600 Apple display for example), and it’d be great to not be restricted to only those old GPUs to gain that capability. For older machines even cleanly driving a new 1920x1080 display without a dongle would be a nice upgrade.
- zare_st 3y agoI have a couple of vintage computers. One is my original first computer. While there is a cheap expansion that would allow me to hook up any VGA screen on it, the part of the experience is the screen. I've been hooked on to retro computing a while, and I flatted out somewhat. It all depends on what you want to do. If you want to play games or run software I suggest you take a good deep look at the emulators. I had a task to do with my XT and that was pulling the data out from it. I did some original programming for that purpose. I spent days loading software from the internet to it in minutes which was kind of miraculous. I programmed some graphics demos on it. I upgraded it with an XT-CF, etc. In the end there is no need to keep that machine up and running on a desk somewhere. The best purpose would be aesthetic, because it is really beautiful as a package, but even if it were in mint condition, I wouldn't risk running it for a couple hours daily for an useless purpose. Although it would be nice to have an 80s style terminal displaying current whether, RSS feeds, etc, it's just a bad way to run your historic machine out of working hours.
- Teknoman117 3y agoMy main trouble with the emulators/virtual machines is that late-90s and early-00s Windows/DirectX games are a huge dead spot. There are a ton of titles from this era that just don't work on current Windows, even with dgVoodoo. I just want to be able to comfortably get rid of this Win98 SE / WinXP dual boot box I have lying around...
- egorfine 3y agoWhat kind of command.com they use? I don't remember any with autocomplete.
- noobermin 3y agoSo, this is probably a super nimby moment, but the growing excitement around "retrocomputing" (really just old machines) means a lot of hardware is becoming super expensive. Like good luck finding an IDE harddrive under 500 MB that works, people are pulling them out of working old machines, putting them on ebay for 100s to even 1000 usd, then shipping them in after thought packaging that has them fall apart on their way there. This hobby is fast becoming a rich man's game for speculators and huskers. It's a tragedy really. I feel like efforts like OP are great because they pull pressure off the literally limited stock thereby making the speculators go elsewhere (segasaturn games and the like mostly).
- PaulHoule 3y agoThe retrocomputing community has struggled with making display controllers because it's not so practical to make one with 74xx/54xx parts (which have just a few gates on a chip) I saw an ad for once circa 1978 in Byte Magazine which was a circuit board about as big as an IBM PC expansion card where both sides of the board were packed with chips. Something like that costs about the same today as it did in 1978. Home computers/game consoles of the time mainly had ASIC display controllers but projects like https://www.commanderx16.com/ https://www.commanderx16.com/ don't really have the volume to justify making an ASIC so they wind up using FPGA (like this card) or microcontrollers to function as display controllers. Note the super low-end https://en.wikipedia.org/wiki/ZX80 https://en.wikipedia.org/wiki/ZX80 did not have a video ASIC but instead tricked the microprocessor into functioning as a video controller which meant that it could only show video when it was done thinking, see https://www.tinaja.com/ebooks/cvcb1.pdf https://www.tinaja.com/ebooks/cvcb1.pdf though that technique can be used today to turn a (secondary) microprocessor for a display controller.
- sydbarrett74 3y agoAwesome work, Yeo! Sorry if I'm asking a duplicate question, but have you considered submitting this to Hackaday?
- yeokm1 3y agoThey have already written an article about it :) https://hackaday.com/2023/09/10/upgraded-graphics-gremlin-adds-hdmi-video-to-vintage-pcs/ https://hackaday.com/2023/09/10/upgraded-graphics-gremlin-ad...