4 ms·
> making ISA video cards instead Sorry, I'll disappoint you. I've experience with these times PCs, with integrated video and with discrete, and must say, this
by simne 3y ago
> making ISA video cards instead
Sorry, I'll disappoint you. I've experience with these times PCs, with integrated video and with discrete, and must say, this was killer difference - just same card on ISA bus was magnitudes slower than integrated.
I've remembered benchmark numbers - on ISA VGA usually considered good about 16000 symbols per second, but integrated gives up to 500k. What all these means - on integrated VGA was possible to play software video on 486 PC.
Only after PCI appeared, in very short time appeared affordable PCI SVGA cards with speeds comparable to integrated and started era of computer home video.
And yes, ALL serious PC brands made their own clones of integrated VGA, beginning from 386SX (386 with 286 16-bit bus), but mostly not for speed, but because at that time it was already cheaper than discrete card at scale.
- rasz 3y agoThere was a short window between 1992-1994 where some highly integrated companies like IBM/PackardBell/HP/DELL shipped computers with local bus capable VGA, but offered only ISA slots for expandability. There were custom pre VLB implementations like OPTi Local bus, Gigabyte had its own, ECS another, etc https://www.vogons.org/viewtopic.php?p=460575#p460575 https://www.vogons.org/viewtopic.php?p=460575#p460575 There was also this one off 1991, pre VESA VLB standardization, Intel/Dell joined experiment called JAWS. Dell PowerLine 450DE/2 DGX Graphics Workstation. Instead of using proper VGA chip it incorporated dumb Inmos G332 framebuffer straight on CPU bus. - Andy Grove himself (Intel CEO!) together with DELL VP Charlie Sauer demonstrating at Comdex '91 (October) https://www.youtube.com/watch?v=CwvOeKqXv18&t=292s https://www.youtube.com/watch?v=CwvOeKqXv18&t=292s how 486DX/50 CPU video rendering is faster than any contemporary accelerator. - Computer Chronicles "486 PC's" episode Feb 1992 https://archive.org/details/intel486 https://archive.org/details/intel486 This one has a shot of actual motherboard. - PC Mag Jul 1992 announcement https://books.google.pl/books?id=X4152M1DLygC&pg=PA59&lpg=PA59&dq=Dell+PowerLine+450DE/2+DGX+Graphics+Workstation https://books.google.pl/books?id=X4152M1DLygC&pg=PA59&lpg=PA... $5500! - BYTE Oct 1992 https://archive.org/details/eu_BYTE-1992-10_OCR/page/n269/mode/2up?q=jaws https://archive.org/details/eu_BYTE-1992-10_OCR/page/n269/mo... $4400, did it really went down $1000 in 4 months? :) - NeXTWORLD September 1993 announcement http://blackholeinc.com/Library/93%20Sept.html http://blackholeinc.com/Library/93%20Sept.html - NeXT price list http://www.nextcomputers.org/NeXTfiles/Articles/NeXTWORLD/Tables/93.5.Sept.Feature2.html http://www.nextcomputers.org/NeXTfiles/Articles/NeXTWORLD/Ta... $7500! - Czech wikipedia of all places :o https://cs-m-wikipedia-org.translate.goog/wiki/Dell_PowerLine_450DE/2_DGX?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=wapp https://cs-m-wikipedia-org.translate.goog/wiki/Dell_PowerLin... - mention in the comments on https://virtuallyfun.com/2012/03/20/dell-unix-lives-again/ https://virtuallyfun.com/2012/03/20/dell-unix-lives-again/ - blog of Charlie Sauer, aforementioned DELL VP (of Advanced Development. May 1989 - October 1993) who showed this computer at Comdex 91 with Intel CEO https://notes.technologists.com/notes/2019/07/01/koko-prolonging-jaws/ https://notes.technologists.com/notes/2019/07/01/koko-prolon... https://notes.technologists.com/notes/2019/07/01/koko-exploring-nextstep-486/ https://notes.technologists.com/notes/2019/07/01/koko-explor... https://notes.technologists.com/notes/2019/07/01/koko-exploring-nextstep-486/ https://notes.technologists.com/notes/2019/07/01/koko-explor... https://notes.technologists.com/notes/2019/07/01/koko-sustaining-dell-unix/ https://notes.technologists.com/notes/2019/07/01/koko-sustai... https://notes.technologists.com/notes/2021/01/19/koko-dell-unix-sustainable/ https://notes.technologists.com/notes/2021/01/19/koko-dell-u...
- simne 3y agoWhat you want to say with all of these stuff? ISA was not an option for VIDEO card at early 90s. VESA (VLB) was not an option, because it was LOCAL bus, closely tied to CPU architecture, so 486 VLB was not compatible with machines with ANY other CPUs even if them exist (PCI have drawbacks, but it nearly complete independent from CPU bus).
- simne 3y agoPS production cost is not tied to market price in market economy. This is very typical mistake to think them tied.
- rasz 3y agoIts historical context. >ISA was not an option for VIDEO card at early 90s. Not sure what you mean when >99% of all Video cards shipped in 1990 were ISA. Alternative were just released EISA and failed MCA. Do you mean not viable performance wise? I dont see how ISA capable of ~5MB/s write speed is a bottleneck when updating 320x200@60 takes less than 4MB/s. Even ignoring fancy ISA 2D video accelerators like 8514 or TIGA, here is this https://github.com/mills32/Little-Game-Engine-for-VGA-EGA https://github.com/mills32/Little-Game-Engine-for-VGA-EGA running on 8MHz 8088 XT equipped with ordinary ISA VGA https://www.youtube.com/watch?v=t98OKbYonQI&t=273 https://www.youtube.com/watch?v=t98OKbYonQI&t=273 @11:10 shows Turbo off 4.7MHz. >VESA (VLB) was not an option, because it was LOCAL bus, closely tied to CPU architecture, so 486 VLB was not compatible with machines with ANY other CPUs https://amiga.resource.cx/exp/egs28 https://amiga.resource.cx/exp/egs28 https://amiga.resource.cx/exp/piccolo https://amiga.resource.cx/exp/piccolo https://amiga.resource.cx/exp/piccolosd64 https://amiga.resource.cx/exp/piccolosd64
- simne 3y agoYou comparing apples with carrots. In democracy you sure have rights for such comparing but I have rights to say you are wrong. > I dont see how ISA capable of ~5MB/s write speed is a bottleneck when updating 320x200@60 takes less than 4MB/s This is because you have not learned theory of https://en.wikipedia.org/wiki/Queueing_theory https://en.wikipedia.org/wiki/Queueing_theory This theory said, to have sustained performance of Y you must have capacity at least 4*Y (and in cases of small systems could be not 4 but 10 or even more). If you don't have free capacity, you will periodically suffer bottlenecks. When Amiga appear, it does not have bottlenecks, it was perfect balanced machine for that time. When appear 386 and even 486, and corresponding 68k CPUs, the more and more architecture become bottleneck, but it was not very obvious. But after started shipping of Pentium machines and drop prices on RAM, so 2M or more become affordable, and people seen 700p screen resolution, it was just too late. > https://amiga.resource.cx/exp/piccolosd64 https://amiga.resource.cx/exp/piccolosd64 They said about Zorro III 32-bit bus, which is not ISA, if you cannot see such obvious thing yourself.