VOGONS


Bought these (retro) hardware today

Topic actions

Reply 53040 of 53280, by Ozzuneoj

User metadata
Rank l33t
Rank
l33t
Trashbytes wrote on 2024-05-23, 16:04:
Ozzuneoj wrote on 2024-05-23, 15:46:
Trashbytes wrote on 2024-05-23, 15:11:

For these cards likely the CPU, even with the CPU overclocked these two GPUs would still be waiting on the CPU to feed them data quickly enough, its pretty hard to bottleneck the AGP 8X bus with any CPU built for it. AGP 8x has the same bandwidth as PCIe 8x or roughly 2.1GB a second .. not even the mighty QX6800 can shovel data fast enough to bottleneck that.

I don't think it's accurate to make sweeping generalizations about bottlenecks. Performance is always dependent on the software and settings first, then the hardware. If you're trying to play a graphically intensive game at a high resolution or with anti-aliasing on an HD 3850 + QX6800, the CPU will not be the bottleneck. If you're trying to play Quake II at 1024x768, then sure, the CPU will probably be the limit. Or, if you're trying to run something that is notorious for making poor use of CPU resources (any Elder Scrolls game), then yes, the CPU will be the bottleneck. In general though, the HD3850 is not an overly powerful card in relation to a good Core2 Duo or Quad. It may be newer, but depending on the game it tends to swing about 15-20% faster or slower than an 8800GT, and usually the 8800GTX outpaces it by a pretty significant margin. Those were the high end cards contemporary to the QX6800.

I would love to see some benchmarks that cover this specific setup though! I have an HD3850 and a couple of HD4650 AGP cards, but I don't have a Socket 775 AGP board myself. It would also be interesting to see different CPUs dropped in to see scaling in different situations, as well as some PCI-E GPUs to compare them.

I'm not wrong here the AGP 8X bus will never be a bottleneck, its very unlikely that any CPU GPU combination using that bus will ever be able to fill the 2.1GB a second bandwidth available to it for it to become the bottle neck. Since I was asked a question about the AGP 8x bus being the bottleneck my answer was focused on that, the reality is that there isn't a CPU on that bus that could saturate it. Its one of the biggest reasons these cards really seem rather silly for AGP cards, any game that could take full advantage of them would require a better CPU to do so. Moving to PCIe and a E8600 for instance, but that's outside of the question here as they are AGP cards.

Benchmarks for both these cards that I have seen pretty much confirm this, feel free to go look them up yourself they are an interesting read.

Okay, I think there must have been a misunderstanding. I wasn't talking about the AGP bus bottlenecks in my reply because I was referring to this statement:

For these cards likely the CPU, even with the CPU overclocked these two GPUs would still be waiting on the CPU to feed them data quickly enough

... in saying this, it sounds like you're saying that even an overclocked QX6800 would be the bottleneck on these GPU (period). But you were actually just saying that the GPU and CPU would be free to "talk" as quickly as they wanted because the AGP 8x bus isn't a bottleneck. That is a perfectly reasonable assessment, I just didn't get that out of that particular statement when I read it, which is why I didn't comment on AGP bottlenecks at all.

I think we're on the same page now. 😁

Now for some blitting from the back buffer.

Reply 53041 of 53280, by Trashbytes

User metadata
Rank Oldbie
Rank
Oldbie
Ozzuneoj wrote on 2024-05-23, 19:47:
Okay, I think there must have been a misunderstanding. I wasn't talking about the AGP bus bottlenecks in my reply because I was […]
Show full quote
Trashbytes wrote on 2024-05-23, 16:04:
Ozzuneoj wrote on 2024-05-23, 15:46:

I don't think it's accurate to make sweeping generalizations about bottlenecks. Performance is always dependent on the software and settings first, then the hardware. If you're trying to play a graphically intensive game at a high resolution or with anti-aliasing on an HD 3850 + QX6800, the CPU will not be the bottleneck. If you're trying to play Quake II at 1024x768, then sure, the CPU will probably be the limit. Or, if you're trying to run something that is notorious for making poor use of CPU resources (any Elder Scrolls game), then yes, the CPU will be the bottleneck. In general though, the HD3850 is not an overly powerful card in relation to a good Core2 Duo or Quad. It may be newer, but depending on the game it tends to swing about 15-20% faster or slower than an 8800GT, and usually the 8800GTX outpaces it by a pretty significant margin. Those were the high end cards contemporary to the QX6800.

I would love to see some benchmarks that cover this specific setup though! I have an HD3850 and a couple of HD4650 AGP cards, but I don't have a Socket 775 AGP board myself. It would also be interesting to see different CPUs dropped in to see scaling in different situations, as well as some PCI-E GPUs to compare them.

I'm not wrong here the AGP 8X bus will never be a bottleneck, its very unlikely that any CPU GPU combination using that bus will ever be able to fill the 2.1GB a second bandwidth available to it for it to become the bottle neck. Since I was asked a question about the AGP 8x bus being the bottleneck my answer was focused on that, the reality is that there isn't a CPU on that bus that could saturate it. Its one of the biggest reasons these cards really seem rather silly for AGP cards, any game that could take full advantage of them would require a better CPU to do so. Moving to PCIe and a E8600 for instance, but that's outside of the question here as they are AGP cards.

Benchmarks for both these cards that I have seen pretty much confirm this, feel free to go look them up yourself they are an interesting read.

Okay, I think there must have been a misunderstanding. I wasn't talking about the AGP bus bottlenecks in my reply because I was referring to this statement:

For these cards likely the CPU, even with the CPU overclocked these two GPUs would still be waiting on the CPU to feed them data quickly enough

... in saying this, it sounds like you're saying that even an overclocked QX6800 would be the bottleneck on these GPU (period). But you were actually just saying that the GPU and CPU would be free to "talk" as quickly as they wanted because the AGP 8x bus isn't a bottleneck. That is a perfectly reasonable assessment, I just didn't get that out of that particular statement when I read it, which is why I didn't comment on AGP bottlenecks at all.

I think we're on the same page now. 😁

Yes, people seem to forget that AGP 8X and PCIe 1.0 X16 are the same speed and bandwidth. AGP 8X was a really huge improvement over AGP 4x and AGPs demise was mainly due to other issues not related to bandwidth or speed.

Reply 53042 of 53280, by Trashbytes

User metadata
Rank Oldbie
Rank
Oldbie

Nab this lot of 10 Intel CPUs hoping on the off chance there might have been a unicorn in the 478 models, sadly there wasn't but there was a I5-9400 in there which is a bit surprising to see in a bunch of retro CPUs, the other CPU of note is a nice I7-7700.

Going to keep the I5 and I7 to upgrade a family members PC and will likely sell the rest off.

CPUs.png

Reply 53043 of 53280, by mtest001

User metadata
Rank Member
Rank
Member

I love the i7 7th gen... Beautiful performances.

/me love my P200MMX@225 Mhz + Voodoo Banshee + SB Live! + Sound Canvas SC-55ST = unlimited joy !

Reply 53044 of 53280, by Trashbytes

User metadata
Rank Oldbie
Rank
Oldbie
mtest001 wrote on 2024-05-25, 06:52:

I love the i7 7th gen... Beautiful performances.

yeh, the I5 is headed for the upgrade but the I7, was thinking it would be good in a SFF Mini Itx machine for lounge gaming, not sure what GPU to pair it with but something in the 10 or 20 series may work ok if I can find a low profile card.

Reply 53045 of 53280, by myne

User metadata
Rank Member
Rank
Member
Trashbytes wrote on 2024-05-25, 06:19:

Yes, people seem to forget that AGP 8X and PCIe 1.0 X16 are the same speed and bandwidth. AGP 8X was a really huge improvement over AGP 4x and AGPs demise was mainly due to other issues not related to bandwidth or speed.

Pcie is simply technically superior in ways most people don't care about, and the industry practically sprinted to adopt it.

Agp didn't fail. It was just out classed.

Things I built:
Mechwarrior 2 installer for Windows 10/11 Re: A comprehensive guide to install and play MechWarrior 2 on new versions on Windows.
Dos+Windows 3.11 auto-install iso template (for vmware)
Script to backup Win9x\ME drivers from a working install

Reply 53046 of 53280, by Trashbytes

User metadata
Rank Oldbie
Rank
Oldbie
myne wrote on 2024-05-25, 07:19:
Trashbytes wrote on 2024-05-25, 06:19:

Yes, people seem to forget that AGP 8X and PCIe 1.0 X16 are the same speed and bandwidth. AGP 8X was a really huge improvement over AGP 4x and AGPs demise was mainly due to other issues not related to bandwidth or speed.

Pcie is simply technically superior in ways most people don't care about, and the industry practically sprinted to adopt it.

Agp didn't fail. It was just out classed.

didn't say it failed, rather it died as you said due tot eh industry rushing to the next new shiny thing.

The next version of AGP had planned to get many of the technical advantages PCIe got. My guess is there was a standards war going on behind the scenes and PCIe won the fight .. much like VHS won over Beta Max and Blu Ray won over HD DVD.

Stuff like that happens all the time.

Reply 53047 of 53280, by Hanamichi

User metadata
Rank Member
Rank
Member
Trashbytes wrote on 2024-05-25, 06:19:
Ozzuneoj wrote on 2024-05-23, 19:47:
Okay, I think there must have been a misunderstanding. I wasn't talking about the AGP bus bottlenecks in my reply because I was […]
Show full quote
Trashbytes wrote on 2024-05-23, 16:04:

I'm not wrong here the AGP 8X bus will never be a bottleneck, its very unlikely that any CPU GPU combination using that bus will ever be able to fill the 2.1GB a second bandwidth available to it for it to become the bottle neck. Since I was asked a question about the AGP 8x bus being the bottleneck my answer was focused on that, the reality is that there isn't a CPU on that bus that could saturate it. Its one of the biggest reasons these cards really seem rather silly for AGP cards, any game that could take full advantage of them would require a better CPU to do so. Moving to PCIe and a E8600 for instance, but that's outside of the question here as they are AGP cards.

Benchmarks for both these cards that I have seen pretty much confirm this, feel free to go look them up yourself they are an interesting read.

Okay, I think there must have been a misunderstanding. I wasn't talking about the AGP bus bottlenecks in my reply because I was referring to this statement:

For these cards likely the CPU, even with the CPU overclocked these two GPUs would still be waiting on the CPU to feed them data quickly enough

... in saying this, it sounds like you're saying that even an overclocked QX6800 would be the bottleneck on these GPU (period). But you were actually just saying that the GPU and CPU would be free to "talk" as quickly as they wanted because the AGP 8x bus isn't a bottleneck. That is a perfectly reasonable assessment, I just didn't get that out of that particular statement when I read it, which is why I didn't comment on AGP bottlenecks at all.

I think we're on the same page now. 😁

Yes, people seem to forget that AGP 8X and PCIe 1.0 X16 are the same speed and bandwidth. AGP 8X was a really huge improvement over AGP 4x and AGPs demise was mainly due to other issues not related to bandwidth or speed.

Technical details show they are quite different and definately not the same speed in anyway:
PCIe 1.0 (16x)- Serial, 8-bit/10-bit encoding, bi-directional, 75 Watts power delivery, 4 Gb/s
AGP (8x) - Parallel, isochronous, 42 Watts power delivery, 2.1 GB/s

PCIe came about as consolidation I'd say, the extra power delivery, flexibility and simplicity over PCI-X for the server market was huge.

HD 4870/X48 and X58 benchmarks with DX10 games started to show dual PCIe 1.0 16x bandwidth making a difference in Crossfire.
Certainly P45/X48/X48 chipsets were performing better than a P35 chipset with 8x/4x PCIe lane split.

A HD 3850 is a gen slower so even PCIe 1.0(8x) and AGP(8x) would very likely not show a performance difference

Reply 53048 of 53280, by Hanamichi

User metadata
Rank Member
Rank
Member
cyclone3d wrote on 2024-05-23, 17:00:
Hanamichi wrote on 2024-05-23, 16:21:
Yeah sounds good, I actually really like a PCI S3 Savage 4 for DOS and Unreal compatibility but not a lot of oomph outside those […]
Show full quote
Shponglefan wrote on 2024-05-23, 00:45:

I do have a few PCI cards (including Voodoo3 and FX5500) as a secondary GPU for Win9x and DOS support. Also thinking of tossing in a Voodoo2 and PowerVR for good measure. 😁

They'll be lots of testing to do to see how feasible this system is.

Yeah sounds good, I actually really like a PCI S3 Savage 4 for DOS and Unreal compatibility but not a lot of oomph outside those uses

I actually just got a SAGP 865 478 board finally to complete my P4 ISA build. Getting worried the interest in P4s is rising and the Korean seller was running low on stock 😂 (Find the eBay TaoBao sellers very unrealiable for SBC stuff)

cyclone3d wrote on 2024-05-23, 06:52:

i7 -> PCIe -> Albatron Atop -> AGP Riser -> FX5900 Ultra

I have the same build but got sidetracked 🤣
I also got those risers from the HK moddiy shop

Two issues to overcome - the strap will be PCX FX5900 128 and some work to get a good compatible driver set for early PCX cards

Nah, not using an Albatron Atop though it would be cool to get one of those boards.

I have a PCX 5900 I can use if I want to make a more modern and stupid fast CPU GeForce FX build which will probably happen.

I was so sure it all made sense 😅 my mistake. Then I'm intrigued what the AGP risers are for!

Reply 53049 of 53280, by Trashbytes

User metadata
Rank Oldbie
Rank
Oldbie
Hanamichi wrote on 2024-05-25, 11:00:
Technical details show they are quite different and definately not the same speed in anyway: PCIe 1.0 (16x)- Serial, 8-bit/10-bi […]
Show full quote
Trashbytes wrote on 2024-05-25, 06:19:
Ozzuneoj wrote on 2024-05-23, 19:47:

Okay, I think there must have been a misunderstanding. I wasn't talking about the AGP bus bottlenecks in my reply because I was referring to this statement:

... in saying this, it sounds like you're saying that even an overclocked QX6800 would be the bottleneck on these GPU (period). But you were actually just saying that the GPU and CPU would be free to "talk" as quickly as they wanted because the AGP 8x bus isn't a bottleneck. That is a perfectly reasonable assessment, I just didn't get that out of that particular statement when I read it, which is why I didn't comment on AGP bottlenecks at all.

I think we're on the same page now. 😁

Yes, people seem to forget that AGP 8X and PCIe 1.0 X16 are the same speed and bandwidth. AGP 8X was a really huge improvement over AGP 4x and AGPs demise was mainly due to other issues not related to bandwidth or speed.

Technical details show they are quite different and definately not the same speed in anyway:
PCIe 1.0 (16x)- Serial, 8-bit/10-bit encoding, bi-directional, 75 Watts power delivery, 4 Gb/s
AGP (8x) - Parallel, isochronous, 42 Watts power delivery, 2.1 GB/s

PCIe came about as consolidation I'd say, the extra power delivery, flexibility and simplicity over PCI-X for the server market was huge.

HD 4870/X48 and X58 benchmarks with DX10 games started to show dual PCIe 1.0 16x bandwidth making a difference in Crossfire.
Certainly P45/X48/X48 chipsets were performing better than a P35 chipset with 8x/4x PCIe lane split.

A HD 3850 is a gen slower so even PCIe 1.0(8x) and AGP(8x) would very likely not show a performance difference

Yep PCIe 8x is 2gb/s ..ah well ..gotta love the gotcha.

Reply 53050 of 53280, by Hanamichi

User metadata
Rank Member
Rank
Member
Shponglefan wrote on 2024-05-23, 16:40:
Got one of those too, but still haven't found a build for it yet. As you say, it's got a narrow range of practical use. […]
Show full quote
Hanamichi wrote on 2024-05-23, 16:21:

Yeah sounds good, I actually really like a PCI S3 Savage 4 for DOS and Unreal compatibility but not a lot of oomph outside those uses

Got one of those too, but still haven't found a build for it yet. As you say, it's got a narrow range of practical use.

I actually just got a SAGP 865 478 board finally to complete my P4 ISA build. Getting worried the interest in P4s is rising and the Korean seller was running low on stock 😂 (Find the eBay TaoBao sellers very unrealiable for SBC stuff)

Sounds interesting! Will you be posting a thread on it?

Always love to see more P4 / ISA based builds.

Yes the case requirement is tricky 13 slots!

pciagp13s2lg.jpg

I've purchased one of these from mercari which has modular parts of 3mm aluminium you can swap so it might work with a custom rear io plate:
https://web.archive.org/web/20060208143054/ht … tflex/index.php

SMART FLEX.png

Reply 53051 of 53280, by Karbist

User metadata
Rank Member
Rank
Member

Got a box of six motherboards for 15 bucks. the interesting ones are Asus p3b-f, Epox nforce 2, giga 8vt800, the rest are meh.
will they work or they all are junk. 🤣

A.jpg
B.jpg

Reply 53052 of 53280, by oh2ftu

User metadata
Rank Member
Rank
Member

A Celeron 333 for 2,50eur. + shipping (6e). No idea if it'll work or not 😀

Reply 53053 of 53280, by Trashbytes

User metadata
Rank Oldbie
Rank
Oldbie
oh2ftu wrote on 2024-05-25, 14:14:

A Celeron 333 for 2,50eur. + shipping (6e). No idea if it'll work or not 😀

Slot 1 or Socket 370 type ?

Reply 53054 of 53280, by Jccwu

User metadata
Rank Newbie
Rank
Newbie

I know you love this kind of stuff

Reply 53055 of 53280, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie
Karbist wrote on 2024-05-25, 12:55:
Got a box of six motherboards for 15 bucks. the interesting ones are Asus p3b-f, Epox nforce 2, giga 8vt800, the rest are meh. w […]
Show full quote

Got a box of six motherboards for 15 bucks. the interesting ones are Asus p3b-f, Epox nforce 2, giga 8vt800, the rest are meh.
will they work or they all are junk. 🤣

A.jpg
B.jpg

Violet one is so messed up it's going to work 100% 🤣
Good luck and post back for the curious bunch.

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 53056 of 53280, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

Hah, yeah... I've got a couple of "very pretty" boards looking like they just came out of the box, dead as a doornail, and no clues, and a few really gnarly battered POS that run fine.

Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.

Reply 53057 of 53280, by cyclone3d

User metadata
Rank l33t++
Rank
l33t++
Hanamichi wrote on 2024-05-25, 11:20:
Yes the case requirement is tricky 13 slots! pciagp13s2lg.jpg […]
Show full quote
Shponglefan wrote on 2024-05-23, 16:40:
Got one of those too, but still haven't found a build for it yet. As you say, it's got a narrow range of practical use. […]
Show full quote
Hanamichi wrote on 2024-05-23, 16:21:

Yeah sounds good, I actually really like a PCI S3 Savage 4 for DOS and Unreal compatibility but not a lot of oomph outside those uses

Got one of those too, but still haven't found a build for it yet. As you say, it's got a narrow range of practical use.

I actually just got a SAGP 865 478 board finally to complete my P4 ISA build. Getting worried the interest in P4s is rising and the Korean seller was running low on stock 😂 (Find the eBay TaoBao sellers very unrealiable for SBC stuff)

Sounds interesting! Will you be posting a thread on it?

Always love to see more P4 / ISA based builds.

Yes the case requirement is tricky 13 slots!
pciagp13s2lg.jpg

I've purchased one of these from mercari which has modular parts of 3mm aluminium you can swap so it might work with a custom rear io plate:
https://web.archive.org/web/20060208143054/ht … tflex/index.php
SMART FLEX.png

You will need a rack mount backplane case for that PIAGP backplane but guessing you already know that.

Did the one you ordered come with the PISA board which is a requirement for the ISA slots to function?

Yamaha modified setupds and drivers
Yamaha XG repository
YMF7x4 Guide
Aopen AW744L II SB-LINK

Reply 53058 of 53280, by Hanamichi

User metadata
Rank Member
Rank
Member
cyclone3d wrote on 2024-05-25, 14:46:
Hanamichi wrote on 2024-05-25, 11:20:
Yes the case requirement is tricky 13 slots! pciagp13s2lg.jpg […]
Show full quote
Shponglefan wrote on 2024-05-23, 16:40:

Got one of those too, but still haven't found a build for it yet. As you say, it's got a narrow range of practical use.

Sounds interesting! Will you be posting a thread on it?

Always love to see more P4 / ISA based builds.

Yes the case requirement is tricky 13 slots!
pciagp13s2lg.jpg

I've purchased one of these from mercari which has modular parts of 3mm aluminium you can swap so it might work with a custom rear io plate:
https://web.archive.org/web/20060208143054/ht … tflex/index.php
SMART FLEX.png

You will need a rack mount backplane case for that PIAGP backplane but guessing you already know that.

Did the one you ordered come with the PISA board which is a requirement for the ISA slots to function?

I prefer to come up with something a bit custom rackmount case are usually annoyingly deep

Yeah brought the backplane 7 years ago! Came with something like pisa-kit-01.

Good point though the isa slots are unusable without the pisa part for potential buyers