• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

SickBeast Marks© are the new Rating System of choice!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
7.65M

510x24x625 7800GTX.

I think you have a problem when comparing X1800 with X1900. This benchmark clearly only works for 1:1 pipeline: Pixelshader cards.

Maybe X1900 should multiply with 1.3
 
Maybe it should be: core clockspeed x (Shaders + TMU +ALU + ROP) x memory speed

to compaire cards with 256-bit memory controller.
 
Originally posted by: pol II
My 7800GTX = 6,480,000 Sick Beast Marks (I think that's correct...)

If my 512MB GTX's do a third of that, I'd say you mismathed somewhere. 😛
 
Originally posted by: Ronin
Originally posted by: pol II
My 7800GTX = 6,480,000 Sick Beast Marks (I think that's correct...)

If my 512MB GTX's do a third of that, I'd say you mismathed somewhere. 😛

You're missing a 7 in your score somewheres 😉
 
With my X800 at 540/590, I get 5,097,600 SickBeast Marks©!!!

Therefore, I can outperform a 7800 GT, and conclude that this system is the most accurate ever devised.
 
Originally posted by: Ronin
2729600 SickBeast Marks©

605*24*940*2 = my result

You can't multiply by 2 to get an SLI result.

You need to wait for SickBeast Mark© 2006 version 1.1.

If you're lucky you'll get a developer build. It won't favour nVidia though, and it won't be a part of TWIMTBP. 😛
 
445054.5 - quite crappily as my 9700 mobility has only 4 pipelines and a really slow RAM clock - that might even be doubled there, I don't know.

Anyway, how are you going to deal with a unified shader system if there is no such thing as a "pipeline" anymore??

Also the above noted: RAM interface width, vertext shaders, quantity of ram, mismatch between RAM and core creating bottlenecks (i.e. Geforce4 MX 440 with 512MB 1500MHz ram.. hey, it could happen!) might mess up your system.
 
Problem with that system, have you forgotten about memory totally? a 256MB Card will outpeform a 128MB Card, and a card with a 256-bit memory interface will be faster than a card with a 128bit interface at the same clock speed.
 
Originally posted by: SickBeast
Oh and for the record, my system predicts that the 7900GTX will be 88% the speed of the X1900XT. 🙂

Your system also predicts that an X1900XTX and an X1800XT will have the same performance if I'm not mistaken....
 
Originally posted by: jiffylube1024
Originally posted by: SickBeast
Oh and for the record, my system predicts that the 7900GTX will be 88% the speed of the X1900XT. 🙂

Your system also predicts that an X1900XTX and an X1800XT will have the same performance if I'm not mistaken....
Hell No!!! The 1900 has 48 shaders, therefore it gets three TIMES the beastmarks as an X1800!!!! :laugh:

Oh and I have 4.65 million beatmarks 🙂
 
Originally posted by: XabanakFanatik
Ronin, I'm afraid you mismathed. You get 27,297,600 SickBeast Marks.

My 7800 GTX gets 8,400,000 at 500/1400

Well I'll be damned...must have crossed my eyes when looking at calc LOL.

And who wants a benchmark that is biased to once manufacturer? 😉
 
A somewhat more accurate benchmark would be:

(core clock * (texture units + pixel shaders)/2 + mem clock * (bus width/16)) / 500
I divide by 500 to give the number some meaning

for example:
x1900xtx = (650 * (16+48)/2 + 775 * (256/16)) / 500 = 66.4
7800gtx512 = ((550 * 24) + 850 * (256/16)) / 500 = 52.4
x850xtpe = ((540 * 16) + 590 * (256/16)) / 500 = 36.16
6600gt = ((500 * 8) + 500 * (128/16)) / 500 = 16

This way the numbers are at least somewhat representaive of the cards' performance in some theoretical game at some undetermined settings.

*revised for balance
 
Back
Top