Odds are it's the trident cards there not exactly the best performing cards.

chrismeyer6 wrote:

Odds are it's the trident cards there not exactly the best performing cards.

...unless it's a 9400Cxi. 😀

...unless it's a 9400Cxi.

More like 9440AGi.

When i play, I get the feeling like the game runs at 30fps but on some parts it drops to 20fps

Don't trust your feelings, Luke. Use timedemo command and share your benchmark results with us.

3DBench should give you 65-70FPS if my memory servers me right. In DOS there isn't much difference between standard VLB cards (10% at most).

But also remember that doom has capped frame rate of 35fps when actually playing the game.

Baoran wrote:

But also remember that doom has capped frame rate of 35fps when actually playing the game.

Really? I could swear Doom feels like >60FPS on anything faster than the original Pentium

jheronimus wrote:

Really? I could swear Doom feels like >60FPS on anything faster than the original Pentium

Yea, I know, it feels really smooth.
Probably because of the double-buffered graphics and the subpixel-corrected rendering that give a very stable, tear-free image, and respond to very subtle motion.
35 fps is quite smooth, when done right, I suppose.

chrismeyer6 wrote:

Odds are it's the trident cards there not exactly the best performing cards.

It depends entirely on the controller and how it's used. From a recent 20-card VLB Doom benchmark I did:

• 9400CXi: I have two of these and they are the slowest in my collection my far
• 9440AGi: significantly faster than the 9400CXi
• 9440 (without a suffix): the fastest VLB card I own. Beats the two ET4000/W32P cards and all Cirrus Logic cards I've tried so far

In DOS there isn't much difference between standard VLB cards (10% at most).

The difference in framerate between the 9440 and 9400CXi is about 20% on my machine (in Doom, again).

Run Doom "-timedemo" and you will be astounded how "slow" it really runs.
Especially DoomII.

The difference in framerate between the 9440 and 9400CXi is about 20% on my machine (in Doom, again).

What about 9440 vs 9440AGi?

gloomkeep wrote:

chrismeyer6 wrote:

Odds are it's the trident cards there not exactly the best performing cards.

In DOS there isn't much difference between standard VLB cards (10% at most).

The difference in framerate between the 9440 and 9400CXi is about 20% on my machine (in Doom, again).

Not sure what model is my Trident, but in DOS it's as fast as S3 864 with some variations depending on the benchmark.

But in my book standard VLB cards are Cirrus Logic and S3. Never seen a VLB Trident back in the days. Only ISA cards.

Looks alright to me, these are my results from Phils Benchmark suite with a VLB Mach64
DX2 66, 3dBench2 FPS=43.4, PCPBench FPS=11.1, Doom FPS=26.41, Realticks=2828, Quake FPS=6.8
DX4120, 3dBench2 FPS=52, PCPBench FPS=13.9, Doom FPS=30.64, Realticks=2437, Quake FPS=8.5

today i woke up and saw the exact model i was using is a trident 9400CXI
I really thought that with that card i would get a near perfect experience on doom1 with a pentium overdrive 486 100mhz, but i did not.

Tonight i will try doing the timedemo with the trident and also with a tseng et4000, i got one of those but my tseng model is not the model that is the fastest one. Im wondering if the pentium overdrive is the problem.

gloomkeep wrote:

It depends entirely on the controller and how it's used. From a recent 20-card VLB Doom benchmark I did: […]
Show full quote

chrismeyer6 wrote:

Odds are it's the trident cards there not exactly the best performing cards.

It depends entirely on the controller and how it's used. From a recent 20-card VLB Doom benchmark I did:

• 9400CXi: I have two of these and they are the slowest in my collection my far

That's interesting, because I had a 9400 CXi back in the day and never ran into any of the issues with speed often mentioned on this forum. So I guess not all cards with that chipset are equal...

I would suggest running top bench and looking at the reported speed of your video memory for comparisons in the thread. Lower numbers will produce higher frames in Doom. How fast pixels can get blasted to the video memory will be your bottleneck on that pentium in doom I'd bet

You can have the same card e.g. cirrus logic cl-gd5428 from multiple different vendors and theyll peform much different from eachother due to the choice of memory.

Even a pentium overdrive wont compensate, only bus speed can help. If you can get a cpu running at 50mhz bus (and the video card takes it) you'll get better performance. A dx4 100mhz or better coupled with a low latency vlb card should lock in 35 fps *most of the time in doom.

Edit: im assuming you have working l2 cache without excessive wait states, I believe sometimes the pod can be picky about l2 cache.

BeginnerGuy wrote:

How fast pixels can get blasted to the video memory will be your bottleneck

If only it were a Sega Genesis....

The Serpent Rider wrote:

What about 9440 vs 9440AGi?

The performance of my AGI is roughly in the middle, between the 9440 and 9400CXi.

Testing was done on a Soyo 4SA W2/W5 with X5-133@160, most aggressive memory timings I could achieve.

I'll probably repeat the testing on another mainboard and with different bus speeds (and different card jumper settings), etc to rule out setting differences.

ok here are some doom1 timedemo demo3

trident vlb cxi : 3863 gametics in 4392 realtics

tseng et4000 vlb: 3863 gametics in 3683 realtics

diamond S3 vlb : 3863 gametics in 3803 realtics

i do still see a slight frame drop with tseng and diamond on open areas.

im starting to like more the amd 133mhz with s3 trio64 pci

the CPU is one pentium dx4 Overdrive 100mhz ( the ones that come with a heatsink kind of purple black, that gets hot like CRAZY)

That's 486 DX4 with write through cache.

Benchmark results are more or less normal for DX4 Overdrive with an average cache and memory timings. With AMD 133-160mhz and S3 Trio64V2/DX or Virge you can get around 50+ fps.