I would have loved to see Stewart Cheifet’s reaction if you walked on set with this on a disk. “🤯 ... Well. That’s all the time we have for today. We hope to see you next week, on The Computer Chronicles.”
Ha! That would have been amazing. That reminds me, I forgot to mention in the video that a few very famous people have actually seen it and given really amazing comments. Relying on memory, but I think this included John Romero and Tim Sweeney!
Imagine showing that to a kid in the 90s.. I was 12 or something when I watched second reality on a 386. Watching this on a much older machine would have killed me 😂 Awesome work.
I'd have been flawed if you showed me an iPhone at age 12, but only if you could program it. This is the sort of thing I'd love to have done with my XT when I was a kid, but I certainly didn't have the know how back then.
I remember when I saw Second Reality for the first time. A friend of mine showed it to me and I was blown away (kind of like with this demo 😉). I also remember us having delusions of grandeur and starting our own demo group, which of course never amounted to anything because none of us four (or were we five?) were 'that kind of genius'... Other than a script for our first demo and a couple of tools of dubious quality, nothing ever came of it.
Yeah it sounded pretty amazing. That was one pretty awesome sound system. The high pitched sound in the credits was fortunately not too terrible either. I don't know if the people responsible for the sound system filtered it out to some degree, but in the video itself it is really evident. I don't know how the organisers knew what volume to use and how to EQ the whole thing because as far as I know there was no dress rehearsal for the sound.
@@PCRetroTech Interesting. You know, it might just be the room and robust speakers does the trick. What helped it out a lot was the bit of echo. That takes the sharp, flat edges off the otherwise really bland square waves. It was just nice. And WOW. Abusing CGA can really produce. That was a lot of fun, and I'm very happy to see the early 8 and maybe soon 16 bit PC's get some scene love. I love hardware at that scale. It's the most fun and rewarding and impressive when people squeeze some great art out of it.
@@Rickdens If I recall correctly, the output was taken directly from the speaker connector. The audio had to be clean as the sound system was about 50,000W. It actually shook your clothing it was so powerful!
I remember watching 8088mph for the first time and it was super cool. Found this demo couple of months ago and my jaw dropped and hit the floor and couple of days ago I showed this to my dad and the same happened to him haha, definitely my favorite demo till now!
Technicalities aside, this was a well presented demo. It lures you in and builds you up slowly. 16 Colours at 640 x 200, even I stood up and started clapping, and I'm just sitting at home watching this on UA-cam.
In my entire life, I would have never thought that an amazing show of sheer skill ran on an IBM 5150 at 4.77mhz and only 640kb of RAM would be rendered and displayed to me with an RTX 3060 with 12GB!! of VRAM at 1777mhz
I'm unclear - it says "no composite tricks", but it certainly looks like it has the slight fuzziness of composite - is it actual TTL CGA just run through a video digitizer that gives it composite softness? Would it look this good on an actual 5153 CGA monitor? Edit: Watched the direct capture (on x86VileR's channel) and yep, definitely TTL mode. Wow.
@@Saturn2888 The monitor is capable of 16 colours. It's just a matter of using lots of CRT controller tricks to ensure as many of them are displayed as possible. Of course text mode can display 16 colours at once, so chopping up a text mode is the main idea.
@@Saturn2888 I think a lot of people are curious about that. There is a video on the channel about how I did my part (the Glenz effect), but everyone will have to wait for the writeups that the other guys release to find out how everything else is done. First comes the final version though, which everyone is going to start working on in October I believe (but it could take an indeterminate amount of time to finish).
If I had a time machine, I'd be sorely tempted to take a floppy with this program back to 1981, to the IBM guys who designed the 5150, and see the looks on their faces when they run it.
There are undocumented modes, but I think we don't use any in this particular demo (other than the "undocumented" but actually quite well-known 160x100 mode). There's a hell of a lot of changing stuff on the fly. But there are lots of undocumented tricks there too. There are people on other platforms with the same chip who aren't going to be that surprised by most of it. But anyone who doesn't know much about the 6845 is going to be really surprised by it when the writeups are done, methinks.
@@kijete It depends how you look at it. If any 80 char mode with two pixel high chars and two colours per char is thought of as 160x100, then throughout. But as an actual 160x100 mode with half char wide "pixels", maybe only one effect, e.g. the chessboard (if that even counts). I only worked on one effect extensively, so getting down into the specifics isn't really possible. I'll be just as interested to read the writeups when the final version is done as everybody else will be.
Part of me is curious if, by using interlaced mode on the CRTC, you can get some different bit patterns on the characters that could work better in some cases? At the very least, by using only lines 1 and 3 or 2 and 4 of each character, that could be a low-maintenance 1024-color-on-composite setup right there (U and !! are long enough to still work, and both dithers would line up by skipping every other line). Also could allow, for example, alternating 2 and 3 lines per char so you can go further into overscan or have a scroll buffer without impacting image quallity that much. Making art for that, on the other hand... A small potential bug: Looking up on the heightmap landscape takes longer than I would expect (at least if it's just "scroll the viewport up a couple of chars", like it was usually done), and, if swivieled too much, seems to break everything (though that could be 86Box not emulating REP MUL, if that was used...) Also, the music sync varies depending on when you exit that scene -- if done just right like here, it swaps smoothly onto the credits, but in VileR's recording it doesn't. PS: Good, UA-cam didn't send my comments to the trash can for once! PPS: How could you get the actual original card font onto a text tweakmode for the intro!? Unless I'm missing something, you can't get lines THAT fine in true text, so it has to be a tricky-to-draw-exactly tweakmode...
I've never personally used the true interlaced mode. It's easy to find statements to the effect that it doesn't work at all on the IBM PC. I've heard rumours about people rescuing something from the mode, but I'd have to consult with them to see what is possible. But unless they've already used it in a demo, it'd be a closely guarded secret. Having never seen the interlaced mode work, I'm not even sure what it does. It makes sense that it pulls from odd and even lines of the character for alternate frames. But I'm not sure exactly how to suppress one frame being displayed or whether the lines end up on consecutive rasterlines on the CRT. I also don't know how you'd detect which frame was being displayed at any given time. So at first blush it seems like there are some technical complications. Certainly 86box won't accurately emulate the voxels landscape. As for your PPS, well that's a detail I can't reveal until one of the others does a writeup. Actually, I didn't look into how it was done myself, and I originally thought they just simulated the original font. But obviously there are a few possible ways to think about and I'm actually not sure which was used.
There are certainly some interesting things possible using the interlaced mode on the CRTC. But there are also some complications which make it difficult to set up. We didn't end up using that mode in this demo, but do watch out for future projects... Area 5150 doesn't use REP MUL so it's quite likely that the bug you mentioned is present on real hardware too. The height controls don't just scroll, they rewrite the perspective table which is why they are so slow - there are various optimisations possible if that code ends up in a game or a demo which uses these height changes.
@@PCRetroTech I read the half scan line isn't output on the CGA. It's been a long time, and I'm struggling to remember the source. Back in the day, some of us were wondering about better vertical resolution and how many machines did not seem capable of it.
I searched youtube for some basic info on how they manipulated graphics modes to create the different sections of the demo and couldnt really find much. I believe in your video you mention manipulation of ascii modes. Is it possible for you to give a quick overview of the methods used to bang on the 6845? Would any of the modes be feasible in an application or game or do the techniques used eat too much cpu?
Many of the effects are very CPU intensive and their facility in a game would be doubtful. It would be possible, but it's a huge amount of work. Some of the "modes" do require CPU cycles to remain in the mode, but the amount of time depends on the effect. As for how it is all done, there will be writeups by everyone eventually, and also a final version of the demo. It would not be right for me to talk about anything other than the effect I contributed. Besides that, everyone needs to take a break right now. Producing a demo on this scale is exhausting.
@@PCRetroTech Understood. I did just read about the 6845 chip and discovered many exploitable items such as how the chip reads one line of character at a time repeatedly until the end of the character which could be used to manipulate horizontal resolution effects then a start of new frame poke at any time etc.. so now some things become more clear. Thanks for the reply.
@@PCRetroTech Wow I didnt realize just how programmable that chip is and that other circuits handle color although I should have known about the color portion especially since this chip is used in so many areas and each implementation produces varying (or no) color.
Hi! I downloaded and tested this out on my Tandy 1000EX. It doesnt quite work correctly. Some parts are absolutely fine. Others, not so much. I would provide you with a capture if it's helpful. I'm saying this fully understanding that the target was an IBM 5150. I also read the .nfo file in the archive that says "We intend to release a final version later in the year." Please consider my request to add Tandy support if you're able in the final version?
I think it is highly unlikely this will be ported to Tandy 1000. Having said that, at least the CRTC is loosely based on the MC6845, so it is not completely and totally out of the question that it could possibly in theory be done, at least in part.
@@douro20 Area5150 might do software DRAM refresh near the very end, I forget now. In general the effects at the end are harder to port because they do more and more crazy things for sure.
Amazing. That's stock 5150 CGA too with no sound card or accelerator card / co-processor / NEC V20 etc isn't it? Absolutely phenominal work. I may have to acquire a 5150 again. Just recently sold one after restoring it. Thank you for sharing this with us ❤️
Correct. The machine has 640kb of memory, but it's a stock 5150 (with hard drive for now, although there are people running it with a dual floppy). Final version may run on a single floppy.
there's a hole in the floor under where my jaw used to be
LOL
What a crazy achievement! It took 40 years for the IBM PC 5150 with 4.77MHz i8088 CPU to finally evolve into the ultimate performance mode! Thank you!
I would have loved to see Stewart Cheifet’s reaction if you walked on set with this on a disk.
“🤯 ... Well. That’s all the time we have for today. We hope to see you next week, on The Computer Chronicles.”
Ha! That would have been amazing. That reminds me, I forgot to mention in the video that a few very famous people have actually seen it and given really amazing comments. Relying on memory, but I think this included John Romero and Tim Sweeney!
@@PCRetroTech That's a heck of a compliment. I would say congratulations, but I'm not sure it would stack up in comparison. haha :-D
@@nickwallette6201 Ha ha! Thanks. {quickly Googles Nick Wallette}
@@PCRetroTech LOL I'll save you the time. To anyone but my cat, I'm nobody of consequence. :-D (And to the cat, I'm just Food Guy.)
Imagine showing that to a kid in the 90s.. I was 12 or something when I watched second reality on a 386. Watching this on a much older machine would have killed me 😂
Awesome work.
I'd have been flawed if you showed me an iPhone at age 12, but only if you could program it. This is the sort of thing I'd love to have done with my XT when I was a kid, but I certainly didn't have the know how back then.
I remember when I saw Second Reality for the first time. A friend of mine showed it to me and I was blown away (kind of like with this demo 😉). I also remember us having delusions of grandeur and starting our own demo group, which of course never amounted to anything because none of us four (or were we five?) were 'that kind of genius'... Other than a script for our first demo and a couple of tools of dubious quality, nothing ever came of it.
That little beeper system sounds GREAT in the room with the big speakers and echos.
Yeah it sounded pretty amazing. That was one pretty awesome sound system. The high pitched sound in the credits was fortunately not too terrible either. I don't know if the people responsible for the sound system filtered it out to some degree, but in the video itself it is really evident. I don't know how the organisers knew what volume to use and how to EQ the whole thing because as far as I know there was no dress rehearsal for the sound.
@@PCRetroTech Interesting. You know, it might just be the room and robust speakers does the trick. What helped it out a lot was the bit of echo. That takes the sharp, flat edges off the otherwise really bland square waves. It was just nice.
And WOW. Abusing CGA can really produce. That was a lot of fun, and I'm very happy to see the early 8 and maybe soon 16 bit PC's get some scene love.
I love hardware at that scale. It's the most fun and rewarding and impressive when people squeeze some great art out of it.
beep beep beep beep beep to Beep Beep Beep
@@PCRetroTech was a mic put right near the pc speaker or did you guys put some kind of output jack to allow them to attach to their amps?
@@Rickdens If I recall correctly, the output was taken directly from the speaker connector. The audio had to be clean as the sound system was about 50,000W. It actually shook your clothing it was so powerful!
I remember watching 8088mph for the first time and it was super cool. Found this demo couple of months ago and my jaw dropped and hit the floor and couple of days ago I showed this to my dad and the same happened to him haha, definitely my favorite demo till now!
Thanks, I'm glad you both liked it!
Wow, nobody ever got excited about a 640x200 image at 16 colors but this is great
Well, the card can only do 2 colours at that resolution. :-)
Technicalities aside, this was a well presented demo. It lures you in and builds you up slowly. 16 Colours at 640 x 200, even I stood up and started clapping, and I'm just sitting at home watching this on UA-cam.
In my entire life, I would have never thought that an amazing show of sheer skill ran on an IBM 5150 at 4.77mhz and only 640kb of RAM would be rendered and displayed to me with an RTX 3060 with 12GB!! of VRAM at 1777mhz
Incredible demo!
I'll second that. Wow. Just wow.
Music sounds like the Zak McKracken intro lol, love it.
I like the effect of the black clapping silhouettes on the bottom. How was it achieved?
Ha! Yes, that was the most expensive part. That involved putting actual people on aeroplanes and sending them to Germany.
Impressive for the platform indeed. Nice work kids!
TTL CGA??!! Can't believe it! Impressive!
Yep, absolutely!
This is crazy and all at 4.77MHz
Is it 4.77 ? Or did it just say 8088 and cga.
@@easyerthanyouthink its the last thing on the screen at the end. It says 4.77
Breath Taking....
Amazing!!
Very impressive 🤩
I'm unclear - it says "no composite tricks", but it certainly looks like it has the slight fuzziness of composite - is it actual TTL CGA just run through a video digitizer that gives it composite softness? Would it look this good on an actual 5153 CGA monitor?
Edit: Watched the direct capture (on x86VileR's channel) and yep, definitely TTL mode. Wow.
Yes, definitely. Any softness is just my camera and UA-cam. And I can confirm it looks amazing on a real 5153
@@PCRetroTech how was that done without composite?
@@Saturn2888 The monitor is capable of 16 colours. It's just a matter of using lots of CRT controller tricks to ensure as many of them are displayed as possible. Of course text mode can display 16 colours at once, so chopping up a text mode is the main idea.
@@PCRetroTech are you releasing a video on how you do that? I'm curious how you got such detailed graphics in text mode.
@@Saturn2888 I think a lot of people are curious about that. There is a video on the channel about how I did my part (the Glenz effect), but everyone will have to wait for the writeups that the other guys release to find out how everything else is done. First comes the final version though, which everyone is going to start working on in October I believe (but it could take an indeterminate amount of time to finish).
Pretty amazing.
Great prog!
Keeps me from throwing away my Amiga !
5:57: Is that a Digger reference?
OMG, it's full of Easter eggs!
If I had a time machine, I'd be sorely tempted to take a floppy with this program back to 1981, to the IBM guys who designed the 5150, and see the looks on their faces when they run it.
Wow. How is that even possible. So magic undocumented CGA modes, changing stuff on the fly I am sure,.
There are undocumented modes, but I think we don't use any in this particular demo (other than the "undocumented" but actually quite well-known 160x100 mode). There's a hell of a lot of changing stuff on the fly. But there are lots of undocumented tricks there too. There are people on other platforms with the same chip who aren't going to be that surprised by most of it. But anyone who doesn't know much about the 6845 is going to be really surprised by it when the writeups are done, methinks.
@@PCRetroTech where was the 160x100 mode used??
@@kijete It depends how you look at it. If any 80 char mode with two pixel high chars and two colours per char is thought of as 160x100, then throughout. But as an actual 160x100 mode with half char wide "pixels", maybe only one effect, e.g. the chessboard (if that even counts). I only worked on one effect extensively, so getting down into the specifics isn't really possible. I'll be just as interested to read the writeups when the final version is done as everybody else will be.
Part of me is curious if, by using interlaced mode on the CRTC, you can get some different bit patterns on the characters that could work better in some cases? At the very least, by using only lines 1 and 3 or 2 and 4 of each character, that could be a low-maintenance 1024-color-on-composite setup right there (U and !! are long enough to still work, and both dithers would line up by skipping every other line).
Also could allow, for example, alternating 2 and 3 lines per char so you can go further into overscan or have a scroll buffer without impacting image quallity that much. Making art for that, on the other hand...
A small potential bug: Looking up on the heightmap landscape takes longer than I would expect (at least if it's just "scroll the viewport up a couple of chars", like it was usually done), and, if swivieled too much, seems to break everything (though that could be 86Box not emulating REP MUL, if that was used...)
Also, the music sync varies depending on when you exit that scene -- if done just right like here, it swaps smoothly onto the credits, but in VileR's recording it doesn't.
PS: Good, UA-cam didn't send my comments to the trash can for once!
PPS: How could you get the actual original card font onto a text tweakmode for the intro!? Unless I'm missing something, you can't get lines THAT fine in true text, so it has to be a tricky-to-draw-exactly tweakmode...
I've never personally used the true interlaced mode. It's easy to find statements to the effect that it doesn't work at all on the IBM PC. I've heard rumours about people rescuing something from the mode, but I'd have to consult with them to see what is possible. But unless they've already used it in a demo, it'd be a closely guarded secret.
Having never seen the interlaced mode work, I'm not even sure what it does. It makes sense that it pulls from odd and even lines of the character for alternate frames. But I'm not sure exactly how to suppress one frame being displayed or whether the lines end up on consecutive rasterlines on the CRT. I also don't know how you'd detect which frame was being displayed at any given time. So at first blush it seems like there are some technical complications.
Certainly 86box won't accurately emulate the voxels landscape.
As for your PPS, well that's a detail I can't reveal until one of the others does a writeup. Actually, I didn't look into how it was done myself, and I originally thought they just simulated the original font. But obviously there are a few possible ways to think about and I'm actually not sure which was used.
There are certainly some interesting things possible using the interlaced mode on the CRTC. But there are also some complications which make it difficult to set up. We didn't end up using that mode in this demo, but do watch out for future projects...
Area 5150 doesn't use REP MUL so it's quite likely that the bug you mentioned is present on real hardware too. The height controls don't just scroll, they rewrite the perspective table which is why they are so slow - there are various optimisations possible if that code ends up in a game or a demo which uses these height changes.
@@PCRetroTech I read the half scan line isn't output on the CGA. It's been a long time, and I'm struggling to remember the source. Back in the day, some of us were wondering about better vertical resolution and how many machines did not seem capable of it.
@@andrewmjenner is it 4.77 mhz or.faster ?
@@easyerthanyouthink 4.77MHz
I always love trying to figure out the how - and the only thing I can come up with is borrowing from some super old school c64 techniques?
Очень круто. Если бы это увидели в 1981 году, охренели бы.
Holy Crap, this is mighty impressive!
Это шедевр!!! Невероятно!
Impressive demo on a dinosaur system.
I searched youtube for some basic info on how they manipulated graphics modes to create the different sections of the demo and couldnt really find much. I believe in your video you mention manipulation of ascii modes. Is it possible for you to give a quick overview of the methods used to bang on the 6845? Would any of the modes be feasible in an application or game or do the techniques used eat too much cpu?
Many of the effects are very CPU intensive and their facility in a game would be doubtful. It would be possible, but it's a huge amount of work. Some of the "modes" do require CPU cycles to remain in the mode, but the amount of time depends on the effect.
As for how it is all done, there will be writeups by everyone eventually, and also a final version of the demo. It would not be right for me to talk about anything other than the effect I contributed. Besides that, everyone needs to take a break right now. Producing a demo on this scale is exhausting.
@@PCRetroTech Understood. I did just read about the 6845 chip and discovered many exploitable items such as how the chip reads one line of character at a time repeatedly until the end of the character which could be used to manipulate horizontal resolution effects then a start of new frame poke at any time etc.. so now some things become more clear. Thanks for the reply.
@@PCRetroTech Wow I didnt realize just how programmable that chip is and that other circuits handle color although I should have known about the color portion especially since this chip is used in so many areas and each implementation produces varying (or no) color.
Guys, why only now? It should've been published like 40 years ago… 😁
Had a few things on. Just wait to see what we have for 2064. 🙂
@@PCRetroTech OK, I'm patient. 😄
@@Pixelschillnobrain I'm aware of this - was just kiddin' (although really it would be handy to know this all that 40 years ago…).
Maybe I need to see if my 5155 is repairable.
Always a fun project, even when there isn't a new demo to try. 🙂
good, c-64 level
respect!
Breaking emulators, again??
Absolutely!
Unbelievable
Hi! I downloaded and tested this out on my Tandy 1000EX. It doesnt quite work correctly. Some parts are absolutely fine. Others, not so much.
I would provide you with a capture if it's helpful.
I'm saying this fully understanding that the target was an IBM 5150.
I also read the .nfo file in the archive that says "We intend to release a final version later in the year." Please consider my request to add Tandy support if you're able in the final version?
I think it is highly unlikely this will be ported to Tandy 1000. Having said that, at least the CRTC is loosely based on the MC6845, so it is not completely and totally out of the question that it could possibly in theory be done, at least in part.
@@PCRetroTech Isn't it also highly sensitive to memory timings as it does software DRAM refresh?
@@douro20 Area5150 might do software DRAM refresh near the very end, I forget now. In general the effects at the end are harder to port because they do more and more crazy things for sure.
this is what masters do
Amazing. That's stock 5150 CGA too with no sound card or accelerator card / co-processor / NEC V20 etc isn't it?
Absolutely phenominal work. I may have to acquire a 5150 again. Just recently sold one after restoring it.
Thank you for sharing this with us ❤️
Correct. The machine has 640kb of memory, but it's a stock 5150 (with hard drive for now, although there are people running it with a dual floppy). Final version may run on a single floppy.
N i c e
RoccoW!
Is the demo exe on the net ?
You can download the demo from any good demo outlet, like Pouet, Scene etc.
@@PCRetroTech yep thanks, i know it 👍
Rediculas!