Upgraded my Windows 98 overkill machine (850XT PE) from a AMD 64 X2 6000+ (3GHz) to a AMD Phenom II X4 955 (3.2GHz). Quake 3 went from 401fps to 520fps. 3DMark99 went from 7500 to 7900. I’m sure I’ve seen benchmarks with like 20,000+ so maybe I need some sort of 3DMark99 patch?

zuldan wrote on Yesterday, 09:57:

Upgraded my Windows 98 overkill machine (850XT PE) from a AMD 64 X2 6000+ (3GHz) to a AMD Phenom II X4 955 (3.2GHz). Quake 3 went from 401fps to 520fps. 3DMark99 went from 7500 to 7900. I’m sure I’ve seen benchmarks with like 20,000+ so maybe I need some sort of 3DMark99 patch?

Yeah, that 3DMark 99 score seems way too low.

I got 41615 points on my C2D E8600 + X800 GTO rig. Full benchmarks are in this post.

Joseph_Joestar wrote on Yesterday, 10:23:

zuldan wrote on Yesterday, 09:57:

Upgraded my Windows 98 overkill machine (850XT PE) from a AMD 64 X2 6000+ (3GHz) to a AMD Phenom II X4 955 (3.2GHz). Quake 3 went from 401fps to 520fps. 3DMark99 went from 7500 to 7900. I’m sure I’ve seen benchmarks with like 20,000+ so maybe I need some sort of 3DMark99 patch?

Yeah, that 3DMark 99 score seems way too low.

I got 41615 points on my C2D E8600 + X800 GTO rig. Full benchmarks are in this post.

Very nice scores! i’m also using ATi Catalyst 6.2 reference drivers. I changed the motherboard and still getting the same score so it must be the 3DMark99 build I’m using. Do you have a link to the exact build you’re using?

It's probably because of the deprecation of 3dnow and/or a slower internal method emulating "obsolete" instructions like MMX/SSE.

Eg Iirc mmx and sse allowed 4 numbers to be operated on in parallel with olny one command to do so.

Eg "add 1 & 2, 3&4, 5&6, 7&8" would result in 3,7,11,15

If that instruction lost its optimised path, it could still exist as a valid instruction but take 4x longer (in clock cycles). Which hardly anyone would notice given the last software written for it was running on hardware at least 1/4 of the speed.

"3DMark 99 MAX is the first benchmark ever that has been optimized to fully support both Intel Pentium III SIMD and AMD 3DNow! instructions at the engine level."

myne wrote on Yesterday, 10:45:

It's probably because of the deprecation of 3dnow and/or a slower internal method emulating "obsolete" instructions like MMX/SSE […]
Show full quote

It's probably because of the deprecation of 3dnow and/or a slower internal method emulating "obsolete" instructions like MMX/SSE.

Eg Iirc mmx and sse allowed 4 numbers to be operated on in parallel with olny one command to do so.

Eg "add 1 & 2, 3&4, 5&6, 7&8" would result in 3,7,11,15

If that instruction lost its optimised path, it could still exist as a valid instruction but take 4x longer (in clock cycles). Which hardly anyone would notice given the last software written for it was running on hardware at least 1/4 of the speed.

"3DMark 99 MAX is the first benchmark ever that has been optimized to fully support both Intel Pentium III SIMD and AMD 3DNow! instructions at the engine level."

That makes sense. I think I’ll try 3DMark 2001 tomorrow and see if there is a major boost.

zuldan wrote on Yesterday, 10:42:

Very nice scores! i’m also using ATi Catalyst 6.2 reference drivers. I changed the motherboard and still getting the same score so it must be the 3DMark99 build I’m using. Do you have a link to the exact build you’re using?

I think I got the whole 3DMark pack from here on Vogonsdrivers, but it's been so long I can't be sure.

They are also available on the Futuremark website of course.

zuldan wrote on Yesterday, 09:57:

Upgraded my Windows 98 overkill machine (850XT PE) from a AMD 64 X2 6000+ (3GHz) to a AMD Phenom II X4 955 (3.2GHz). Quake 3 went from 401fps to 520fps. 3DMark99 went from 7500 to 7900. I’m sure I’ve seen benchmarks with like 20,000+ so maybe I need some sort of 3DMark99 patch?

the quake 3 scores look good, the 3dmark99 scores are low you probably need the patch for 3dmark99 to unlock the framerate otherwise the benchmark will be locked at 75fps, without the patch my fx5200 was scoring around 6-7000 with the patch its around 12000, the 850xt should do at least double that depending on cpu.

Joseph_Joestar wrote on Yesterday, 10:56:

zuldan wrote on Yesterday, 10:42:

Very nice scores! i’m also using ATi Catalyst 6.2 reference drivers. I changed the motherboard and still getting the same score so it must be the 3DMark99 build I’m using. Do you have a link to the exact build you’re using?

I think I got the whole 3DMark pack from here on Vogonsdrivers, but it's been so long I can't be sure.

They are also available on the Futuremark website of course.

Thanks for the info. I’m using the build from the 3DMark website.

DudeFace wrote on Yesterday, 11:22:

zuldan wrote on Yesterday, 09:57:

Upgraded my Windows 98 overkill machine (850XT PE) from a AMD 64 X2 6000+ (3GHz) to a AMD Phenom II X4 955 (3.2GHz). Quake 3 went from 401fps to 520fps. 3DMark99 went from 7500 to 7900. I’m sure I’ve seen benchmarks with like 20,000+ so maybe I need some sort of 3DMark99 patch?

the quake 3 scores look good, the 3dmark99 scores are low you probably need the patch for 3dmark99 to unlock the framerate otherwise the benchmark will be locked at 75fps, without the patch my fx5200 was scoring around 6-7000 with the patch its around 12000, the 850xt should do at least double that depending on cpu.

Yes my frames are limited to 75fps for each test. Wish I could find this magical 3DMark99 patch. I’ll try the Vogons 3DMark99 files Joseph linked.

zuldan wrote on Yesterday, 11:28:

Thanks for the info. I’m using the build from the 3DMark website.

One more tip. Be sure to fully disable V-Sync in the Catalyst Control Center before performing any benchmarks. Both for Direct3D and OpenGL. Don't leave it at "Default" or "Auto" etc. Make sure it's actually turned off.

From what I recall, some earlier versions of 3DMark are affected by this. And certain OpenGL games might also report inaccurate benchmark results unless V-Sync is off in the driver panel.

zuldan wrote on Yesterday, 11:30:

DudeFace wrote on Yesterday, 11:22:

zuldan wrote on Yesterday, 09:57:

Upgraded my Windows 98 overkill machine (850XT PE) from a AMD 64 X2 6000+ (3GHz) to a AMD Phenom II X4 955 (3.2GHz). Quake 3 went from 401fps to 520fps. 3DMark99 went from 7500 to 7900. I’m sure I’ve seen benchmarks with like 20,000+ so maybe I need some sort of 3DMark99 patch?

the quake 3 scores look good, the 3dmark99 scores are low you probably need the patch for 3dmark99 to unlock the framerate otherwise the benchmark will be locked at 75fps, without the patch my fx5200 was scoring around 6-7000 with the patch its around 12000, the 850xt should do at least double that depending on cpu.

Yes my frames are limited to 75fps for each test. Wish I could find this magical 3DMark99 patch. I’ll try the Vogons 3DMark99 files Joseph linked.

https://drive.google.com/file/d/1RBV2-D3E5vUk … ?usp=drive_link
The link contains all the patches for retro 3DMarks.
All of them can be configured on a modern machine. For example, 3DM99 on RyZen + QFX3500

Phenom-II-X2 + HD6870

DudeFace wrote on 2025-08-05, 18:28:

nice ive been wanting a mega cd for my mega drive 1 since i was like 7yo, now 32yrs on i still dont own one. lol would also like […]
Show full quote

PTherapist wrote on 2025-07-28, 14:57:

Spent hours over the past couple of days trying to fix a Sega Mega CD Model 1. […]
Show full quote

Spent hours over the past couple of days trying to fix a Sega Mega CD Model 1.

It needed a replacement laser but no matter how much I calibrated the new laser, it wouldn't reliably read CD-R backup discs. Original discs were no problem, but CD-R was so hit and miss. Games would either not boot, or boot fine and develop glitches later. I got it to a point where it would read some backups temperamentally, then I decided to just give up and reassemble the console.

After putting everything back together I thought I'd just test it one more time - only to discover that it's now working perfectly fine and reading all discs. WTF!!!!

The only thing I can think of here is that the issue may have been RF interference between the Mega CD & the Mega Drive, which impacts the much noisier CD-R media more heavily. Though I've never seen anybody reference this as an issue before and the consensus always seemed to be that the shielding was irrelevant.

Either way, once the console is reassembled, the internal metal shielding and external shielding between the Mega CD & Mega Drive is now present. Prior to this, I was testing it out without both sets of shielding, with just the Mega Drive sitting awkwardly on top, directly above the disc drive.

Either that, or the console is just cursed. 🤣

nice ive been wanting a mega cd for my mega drive 1 since i was like 7yo, now 32yrs on i still dont own one. 🤣 would also like a multimega and a mega pc, thought looking at current prices prob should have bought them years ago, anyway the problems you've been having with the cdrs is probably down to the speed you're burning at, i had the exact same issues with my sega saturn and dreamcast cdrs, with the saturn the games would be jumpy and just not load right, i sold my saturn (this was before ode's) then found i had the same issues with dreamcast, i eventually realised it was down to burn speed,

it was 50/50 whether the console would recognise the disc or not and 50/50 if it would detect it as a game or audio cd, when the games did load they were jumpy as shit and just not working right, at first i was burning at 12x since it was the lowest my drive would do, i then found a drive that did 10x, it was better but still unplayable, i then found a drive that will do 8x and since then every disc loads no problem and plays perfectly, im also using cheap as shit discs like £3 for a 10 pack.

i know ps2 dvdrs need to be burned at 4x, xbox360 at 2.4x, or you'll end up with the same problems you mentioned, so for cdr games 8x is probably the maximum you want to go, since mega cd is older you may need to burn lower, also good luck finding a dvd drive that burns that low, you'll probably haver more luck with a cd writer.

also i heard the metal plate is to shield the megadrive from the mega cd's radiation, or shield the mega cd from the megadrives radiation i forget.

In this particular case, it's definitely the console being the issue here, rather than burn speed.

I burn discs at 4x, which is the lowest the burner I use will go. I've got several other burners that won't go lower than 10x-16x, so I'm quite lucky with this old circa-2012 Pioneer CD/DVD Writer/Blu-Ray ROM drive.

A good bit of advice if you ever decide to get a Mega CD, probably go with the Model 2 if you want minimal maintenance. If you fancy a more thorough full restoration project, get the Model 1 like I did and enjoy the pain of a lot of recapping and drive repair (even worse if you get a Model 1 with the Sony disc drive/laser, they're even worse). 🤣

Uploaded a couple of backup/restore cd images in mdf/mds format for time computers onto the vogons library
https://www.vogonsdrivers.com/getfile.php?fil … 2&menustate=0,0

zuldan wrote on Yesterday, 09:57:

Upgraded my Windows 98 overkill machine (850XT PE) from a AMD 64 X2 6000+ (3GHz) to a AMD Phenom II X4 955 (3.2GHz). Quake 3 went from 401fps to 520fps. 3DMark99 went from 7500 to 7900. I’m sure I’ve seen benchmarks with like 20,000+ so maybe I need some sort of 3DMark99 patch?

Make sure vsync is off! If the framerate counter is maxed at like 60 fps while it's running then vsync is still on.

PTherapist wrote on Yesterday, 15:38:

In this particular case, it's definitely the console being the issue here, rather than burn speed. […]
Show full quote

DudeFace wrote on 2025-08-05, 18:28:

nice ive been wanting a mega cd for my mega drive 1 since i was like 7yo, now 32yrs on i still dont own one. lol would also like […]
Show full quote

PTherapist wrote on 2025-07-28, 14:57:

Spent hours over the past couple of days trying to fix a Sega Mega CD Model 1. […]
Show full quote

Spent hours over the past couple of days trying to fix a Sega Mega CD Model 1.

It needed a replacement laser but no matter how much I calibrated the new laser, it wouldn't reliably read CD-R backup discs. Original discs were no problem, but CD-R was so hit and miss. Games would either not boot, or boot fine and develop glitches later. I got it to a point where it would read some backups temperamentally, then I decided to just give up and reassemble the console.

After putting everything back together I thought I'd just test it one more time - only to discover that it's now working perfectly fine and reading all discs. WTF!!!!

The only thing I can think of here is that the issue may have been RF interference between the Mega CD & the Mega Drive, which impacts the much noisier CD-R media more heavily. Though I've never seen anybody reference this as an issue before and the consensus always seemed to be that the shielding was irrelevant.

Either way, once the console is reassembled, the internal metal shielding and external shielding between the Mega CD & Mega Drive is now present. Prior to this, I was testing it out without both sets of shielding, with just the Mega Drive sitting awkwardly on top, directly above the disc drive.

Either that, or the console is just cursed. 🤣

nice ive been wanting a mega cd for my mega drive 1 since i was like 7yo, now 32yrs on i still dont own one. 🤣 would also like a multimega and a mega pc, thought looking at current prices prob should have bought them years ago, anyway the problems you've been having with the cdrs is probably down to the speed you're burning at, i had the exact same issues with my sega saturn and dreamcast cdrs, with the saturn the games would be jumpy and just not load right, i sold my saturn (this was before ode's) then found i had the same issues with dreamcast, i eventually realised it was down to burn speed,

it was 50/50 whether the console would recognise the disc or not and 50/50 if it would detect it as a game or audio cd, when the games did load they were jumpy as shit and just not working right, at first i was burning at 12x since it was the lowest my drive would do, i then found a drive that did 10x, it was better but still unplayable, i then found a drive that will do 8x and since then every disc loads no problem and plays perfectly, im also using cheap as shit discs like £3 for a 10 pack.

i know ps2 dvdrs need to be burned at 4x, xbox360 at 2.4x, or you'll end up with the same problems you mentioned, so for cdr games 8x is probably the maximum you want to go, since mega cd is older you may need to burn lower, also good luck finding a dvd drive that burns that low, you'll probably haver more luck with a cd writer.

also i heard the metal plate is to shield the megadrive from the mega cd's radiation, or shield the mega cd from the megadrives radiation i forget.

In this particular case, it's definitely the console being the issue here, rather than burn speed.

I burn discs at 4x, which is the lowest the burner I use will go. I've got several other burners that won't go lower than 10x-16x, so I'm quite lucky with this old circa-2012 Pioneer CD/DVD Writer/Blu-Ray ROM drive.

A good bit of advice if you ever decide to get a Mega CD, probably go with the Model 2 if you want minimal maintenance. If you fancy a more thorough full restoration project, get the Model 1 like I did and enjoy the pain of a lot of recapping and drive repair (even worse if you get a Model 1 with the Sony disc drive/laser, they're even worse). 🤣

if you're burning at x4 then yeah its gotta be the console, you're lucky to find a drive that burns that low, a mega cd2 is probably the best for reliability, im going for nostalgia and looks over reliability so it will be a model 1 i go for also to match my megadrive 1, also hoping to find a boxed one for that new feeling, looking at prices on CeX its £375, unboxed is about £200 which is more reasonable, a boxed multimega is £850 and unboxed is still over £500, i remember they retailed for either £299 or £399, as for replacing caps thats no problem if the laser is dead ive heard about an ode option.

Joseph_Joestar wrote on Yesterday, 11:31:

zuldan wrote on Yesterday, 11:28:

Thanks for the info. I’m using the build from the 3DMark website.

One more tip. Be sure to fully disable V-Sync in the Catalyst Control Center before performing any benchmarks. Both for Direct3D and OpenGL. Don't leave it at "Default" or "Auto" etc. Make sure it's actually turned off.

From what I recall, some earlier versions of 3DMark are affected by this. And certain OpenGL games might also report inaccurate benchmark results unless V-Sync is off in the driver panel.

Thanks Joseph and pete8475 and everyone else that responded. Silly me had v-sync on. I assumed it was off by default.

Not bad for a Windows 98 machine. Not as insane as DarthSun 98 machine, although I should be able to play a lot more older 98 games 😉