Albatron Ti4200/3D Mark Scores

[shulse]

I don't get it. I recently purchased an Albatron Ti4200 Turbo 128MB vid card. Everywhere on the web I have read that people are getting 10-11,000 on their 3D Mark 2001 SE scores. I have a P4 1.7 processor, 512 of SD RAM and I am only getting 7300 on my scores.

http://www6.tomshardware.com/graphic/20021218/vgacharts-05.html

I have my settings in 3D mark set at the exact same as the person that ran the above tests. He got 11,754. I don't know the internal temp of my case, but I don't think there is anything going on in there that would make me lose 2,500-3,000 in score. Also, my card is already somewhat overclocked compared with regular Ti4200 settings. It has heatsinks on the RAM and a huge cooling fan.

Can anyone offer some suggestions? Oh - this is my second card...I returned the first one because I was unhappy with the scores. With the second card I am getting those same scores.

 

[jumpy]

Except that you are using SDRAM and a 1.7 willamette(sp?) and the other system is using a 3.06 and DDR333. That is why the score is that different.

 

[shulse]

Originally posted by jumpy
Except that you are using SDRAM and a 1.7 willamette(sp?) and the other system is using a 3.06 and DDR333. That is why the score is that different.

It would really make that MUCH of a difference? Seems like such a disparity.

 

[Sazar]

Originally posted by shulse
It would really make that MUCH of a difference? Seems like such a disparity.

his cpu technically has 2 times the performance of your machine... (actually more like 2.5 - 3 times raw performance but lets keep it simple )

also the memory DOES make a difference...

for your system the scores you are getting are fine...

btw 3dmark is a highly synthetic score your real-world gaming performance is NOT dependent on 3dmarks

I will not get into the whole nvidia driver's optimized especially for 3dmark thing might start a flame war...

 

[shulse]

Originally posted by Sazar
his cpu technically has 2 times the performance of your machine... (actually more like 2.5 - 3 times raw performance but lets keep it simple )

also the memory DOES make a difference...

for your system the scores you are getting are fine...

btw 3dmark is a highly synthetic score your real-world gaming performance is NOT dependent on 3dmarks

I will not get into the whole nvidia driver's optimized especially for 3dmark thing might start a flame war...

Ok...ok. Just a little disappointed. Makes sense...to tell you the truth I didn't realize his CPU was that kick-ass.

Let me ask you this....would the CPU and RAM have an effect on the cards overclockability? I assume it would, but again, the disparity seems to be so great. I have my card overclocked and running stable at 280/630. That is it for this card on my machine. A ton of people on newegg's website (where I got the card) are all pretty much saying that they are getting 300/690 stable. Unfortunately not everyone is listing their specs but from the ones that are I do see that their systems do have a better cpu/memory...but not much better.

Thanks for the info guys
:happy:

 

[Sazar]

Originally posted by shulse
Ok...ok. Just a little disappointed. Makes sense...to tell you the truth I didn't realize his CPU was that kick-ass.

Let me ask you this....would the CPU and RAM have an effect on the cards overclockability? I assume it would, but again, the disparity seems to be so great. I have my card overclocked and running stable at 280/630. That is it for this card on my machine. A ton of people on newegg's website (where I got the card) are all pretty much saying that they are getting 300/690 stable. Unfortunately not everyone is listing their specs but from the ones that are I do see that their systems do have a better cpu/memory...but not much better.

Thanks for the info guys
:happy:

some people BUILD their systems to be able to max out the performance for each component... solely for oc'ing you see

the thing to remember is that you are just as likely to blow up a component as you are to gain a couple of fps here and there...

if you are not sure on how something is done than it is better to readup about it first rather than jump in headfirst... that would be my advice....

learn about your system and if it can be optimized any WITHOUT overclocking and THEN consider oc'ing it...

 

[Goatman]

I'd say you should upgrade your mobo and RAM and goto DDR RAM

Asus P4B266-C- $59

512MB Kingston DDR PC2700 RAM- $103.50

That's $162 for a upgrade that would yield you a large performance increase.

 

[jumpy]

Goatman - that upgrade would depend on whether the 1.7 p4 is 423 pin or 478. Shulse - An upgrade of mobo and memory would be the most beneficial upgrade for your system. Maybe the p4b533 to ensure it can run 533 bus p4's for future upgrades.

 

[shulse]

Originally posted by Goatman
I'd say you should upgrade your mobo and RAM and goto DDR RAM

Asus P4B266-C- $59

512MB Kingston DDR PC2700 RAM- $103.50

That's $162 for a upgrade that would yield you a large performance increase.

That makes the most sense. I see now, and agree, it will provide the most bang for the buck.

Learn something every day. That is why I like this forum.

 

[shulse]

Goatman or Jumpy -

I have a last question. I have been looking into the upgrades that you guys were talking about and decided to do it.

I just want to make sure that you aren't actually talking about a processor upgrade. You are talking about moving the cpu from my current motherboard to a newly aquired one, correct? This would make sense with Jumpy's concern about pins and future upgrades.

Also, I really don't see a difference between Goatman's and Jumpy's motherboards. Goatman - does your motherboard only support 478 pins? Jumpy - I am assuming your motherboard supports both?