3DMark2001/03/05 .: vs :. 512MB/1024MB/2048MB - Comparison

Hey,

I have decided to spent few hours tonight benching Futuremark's 3D benchmark against different RAM configurations. I hope that there will be quite a few of you guys that will find this thread interesting. For some it might not be anything new but maybe for many others it will :D So, here we go...

Starring:

2 x 256MB Mushkin PC3200 LvL2 (Winbond BH-5)
2 x 256MB G.Skill PC4800LA (Samsung TCCD)
2 x 512MB G.Skill PC4400LE (Samsung TCCD)
2 x 1GB Corsair TWINX2048-4400PRO (Infineon CE-5)
2 x 1GB G.Skill PC4000HZ (Samsung UCCC)

7800GTX 512MB @ 550/1700 (stock), Forceware 81.98, High Quality
(Property of a buddy of mine, waiting to be vmodded...)

AMD Opteron 150
(One of njkid32's toys...)

I chose CPU speed of 3190Mhz (290x11) as this is the MAX 3D stable frequency for UCCC and TCCD. All benchmarks were run in EXACTLY same way: 3DMark2001 first, 3DM05 second, then reboot and 3DM03 last. In comparo between 512MB TCCD, 1GB TCCD and 2GB UCCC @ 8-4-4-3.0, I think I managed to use absolutely same latencies and other timmings for both. That example is really useful to see how amount of memory affects (if any) score is 3D benchmarks.

2x256MB BH-5

3DMark2001 and 05 @ 266x12=3192MHz DDR532 5-2-2-2.0-1T --- 36748 / 9795

3DMark03 @ 266x12=3192MHz DDR532 5-2-2-2.0-1T --- 20894


2x256MB TCCD

3DMark2001 and 05 @ 290x12=3190MHz DDR580 8-4-4-3.0-1T --- 35868 / 9789

3DMark03 @ 290x12=3190MHz DDR580 8-4-4-3.0-1T --- 20748

Unfortunately, I could not get this pair to run @ DDR580 6-3-3-2.5-1T stable enough to pass all benches...


2x512MB TCCD

3DMark2001 and 05 @ 290x12=3190MHz DDR580 8-4-4-3.0-1T --- 36289 / 9785

3DMark03 @ 290x12=3190MHz DDR580 8-4-4-3.0-1T --- 20846

3DMark2001 and 05 @ 290x12=3190MHz DDR580 6-3-3-2.5-1T --- 36900 / 9812

3DMark03 @ 290x12=3190MHz DDR580 6-3-3-2.5-1T --- 20969


2x1024MB UCCC

3DMark2001 and 05 @ 290x12=3190MHz DDR580 8-4-4-3.0-1T --- 36261 / 9814

3DMark03 @ 290x12=3190MHz DDR580 8-4-4-3.0-1T --- 20895


2x1024MB Infineon CE-5

3DMark2001 and 05 @ 266x12=3192MHz DDR532 5-2-3-3.0-1T --- 36396 / 9804

3DMark03 @ 266x12=3192MHz DDR532 5-2-3-3.0-1T --- 20901


So there you go... :) The only RAM that is missing from this comparison would be 2x512MB BH-5 and 2x1024MB Micron -5B D. Unfortunately, I don't have those sets :( But I think we could expect 2x512MB BH-5 to score better compared to 2x256MB BH-5 due to Bank Interleave kicking in. A config of 2x512MB BH-5 could very possibly outperform TCCD as well, but not by very much. But I would love to see, how 290MHz 8-3-3-3.0 (Crucial Ballistix) compares to Infineon CE-5 @ 266Mhz CE-5 and TCCD @ 290MHz 6-3-3-2.5-1T. If I manage to get one of those I will update this post ;)

Main goal of this comparison is see what differenences are between 1GB and 2GB configs... All of the scores can be bettered by further "tweaking" of some latencies but I chose not to so this time. I just wanted to show the trend :)

I will leave conclusions to you guys ;)

Thanks for stopping by :up:

EDIT: I'm not sure which section is better for this thread: X3DTeam or XB...
So, at some point I might move it or not :D Thanks for understanding...

Thanks for your time and effort bachus_anonym and .....have a happy new year.....:D :toast:

nice results - pretty much what i expected the end results to be :)

WOOT the 1st thread of 2006 for bachus
NICE result!!!! thx

4x512 bh-5 & TCCD for some 2T ? :D

How about a chart in excel? Just a suggestion :)

Thanks, nice results and I also believe 2x512mb bh5/utt type would win this :)

nice results :)

very interseting, i agree a graph would be good!!

That is interesting. I didn't think 2x1024MB would do that well in 05.

Wow... TCCD actually beat out BH5 in 01? Mostly due to bank interleaving I assume... Interesting how interlacing hardly effected 03 at all and didn't do a damn thing for 05.

Thanks for taking to time to do this bachus :)

shouldnt it be 3190 is 11x290 rather than 12x290?

Great work bach! This is the kind of legwork that is so time consuming and costly sometimes. THanx for the comparisons. THis should help out all 3d benchers.

Nice round up :)

Someone send Bach some 2x512 BH-5 to test ;)

Thanks guys!

Yeah, it took few hours of benching :D Damn, I've never shuffled so many sticks and run 3D for that long, in my lifetime :lol: I'm gonna work on a table or chart for it. Now, If only still had 2x512MB BH-5/UTT. I'm gonna have to grab some...

Thanks. This kind of stuff is why I check in to XS daily.

That would be incredible if you could get some 2x512MB BH-5/UTT. Yours ram comparisons are awesome! keep up the good work. Maybe you should make a definitive RAM sticky with all of them + all the databases and such.

great comparison bachus!!!

Quote:

---

Originally Posted by afireinside

Wow... TCCD actually beat out BH5 in 01? Mostly due to bank interleaving I assume... Interesting how interlacing hardly effected 03 at all and didn't do a damn thing for 05.

Thanks for taking to time to do this bachus :)

---

2X 256MB BH-5 would just be slower than 2x 512MB TCCD with tight timings and yes bank interleave helps TCCD keep up :)

great work bachus! :up:

as always, i might add. :)

there is just one thing, 1GB vs 2GB gaming performance goes beyond numbers. benchmarks simply can't show the difference you feel when you play, for example, FEAR on a rig with 2GB RAM... after playing it with 'just' 1GB.

Quote:

---

Originally Posted by bachus_anonym

Thanks guys!

Yeah, it took few hours of benching :D Damn, I've never shuffled so many sticks and run 3D for that long, in my lifetime :lol: I'm gonna work on a table or chart for it. Now, If only still had 2x512MB BH-5/UTT. I'm gonna have to grab some...

---

I still have 2*512MB Corsair BH5 sticks here.
To bad you live the other part of the world.
Nice review anyway, great job. :up:

Quote:

---

Originally Posted by bachus_anonym

Thanks guys!

Yeah, it took few hours of benching :D Damn, I've never shuffled so many sticks and run 3D for that long, in my lifetime :lol: I'm gonna work on a table or chart for it. Now, If only still had 2x512MB BH-5/UTT. I'm gonna have to grab some...

---

How many do you need? :fact:, hmm naa i want to test the sticks first. But it had be nice to ship them to you first. Maybe if i get 2x more i ship them to you before they will be shiped to me.

Quote:

---

Originally Posted by high5

there is just one thing, 1GB vs 2GB gaming performance goes beyond numbers. benchmarks simply can't show the difference you feel when you play, for example, FEAR on a rig with 2GB RAM... after playing it with 'just' 1GB.

---

Agreed! This was just a test I decided to do after I saw a post somewhere claiming that 2GB gives advantage in 3D benching compared to 1GB...

Quote:

---

Originally Posted by Jupiler

I still have 2*512MB Corsair BH5 sticks here.
To bad you live the other part of the world.

---

Hehe, thanks JF :) I had total of 3 sets of 1GB BH-5/UTT but they are all gone now. As soon as I find a good kit I will update this post. Might be very soon, but I don't think I will have that 7800GTX 512MB then. I will say goodbye to it very, very soon :(

Quote:

---

Originally Posted by Lastviking

How many do you need? :fact:, hmm naa i want to test the sticks first. But it had be nice to ship them to you first. Maybe if i get 2x more i ship them to you before they will be shiped to me.

---

That's alright, LV :) I'm sure I will find some good stuff locally... You need them more than I do for your 1M / >293Mhz benching :)

you should do a test with 2 ch-5 sticks too since you did tcc5 and tccd and those are pretty much identical

Great stuff Bachus, very cool of you. Thanks for your time! :toast:

thanks man........ the 01 results are closer than i would have thought !

i've got a pair of bh-5 UTT, but wont be in town until the weekend and the post office isnt open then of course... :rolleyes:

that 2 gig vs 1 gig comment about gaming is true. #'s don't tell it all, but playing frozen throne does, hehe. Bachus, if u pay the shipping, I'll send ya my nice sticks of old old ballistix to test out ;)

seems like a lil 2 many different factors in ur equation to actuallly compare them in the sense of which is better 4 3dmark

@ STEvil and fareastgq

Thanks guys but unfortunately 7800GTX is going to be out the door by Friday the latest :( Thanks for the offer, though!

Quote:

---

Originally Posted by Rayman

seems like a lil 2 many different factors in ur equation to actuallly compare them in the sense of which is better 4 3dmark

---

Can you make it a bit clearer? What do you exactly mean?

Quote:

---

Originally Posted by afireinside

Wow... TCCD actually beat out BH5 in 01?

---

i think not :)
Bachus should lower the timings for his 2x256 bh-5.Im pretty sure he could run these with TRFC 12 and TRTW 1. A lot of 256 sticks are capable doing that @ 266MHz,though dunno about the TCCD chips,just to be a perfect compare.

Quote:

---

Originally Posted by sang

i think not :)
Bachus should lower the timings for his 2x256 bh-5.Im pretty sure he could run these with TRFC 12 and TRTW 1. A lot of 256 sticks are capable doing that @ 266MHz,though dunno about the TCCD chips,just to be a perfect compare.

---

Oops, I just noticed that TRFC was not 12... But you know how much difference it really makes? Really not significant. TRTW=1 @ 266MHz is something that not always can be done and in this case was causing reboots like there was no tommorow :) Even with higher vdimm (those modules are SPi32M stable @ 278MHz).
Also, like I added at the end, TCCD can also be tigtened. All of those runs can be tweaked further, so there is improvement everywhere.
Thing is that 2x512MB TCCD @ 290MHz 6-3-3-2.5-1T will beat 2x256MB BH-5 @ 266MHz 6-2-2-2.0-1T due to Bank Interleave being enabled with 2x512MB. As we know and see from above screenshots, 3DM01 is very sensitive to memory timmings and responds to enabled Bank Interleave very nicely. 2x512MB BH-5 vs 2.512MB TCCD at those clocks could have very results, though.

well,i digged 2 of my old screens for the trfc and trtw issue.With 2x256 Mushkin 3500 L2 those settings are piece of cake. I dont have anymore an A64 rig for testing but with your kindness we can find the exact difference.
Please,make another 2 tests for 3d2001. Find the highest memory frequency for your 2x256 sticks in order to be able to bench with the above timing parameters TRFC 12/TRTW 1 versus TRFC 14/TRTW 2.
Thanks in advance :toast:

Quote:

---

Originally Posted by fareastgq

that 2 gig vs 1 gig comment about gaming is true. #'s don't tell it all, but playing frozen throne does, hehe. Bachus, if u pay the shipping, I'll send ya my nice sticks of old old ballistix to test out ;)

---

wow you noticed a difference with 1gig vs 2gig with a 6600gt in wc3:tft??

i recently stopped playing that game but before i stopped i remembered that game never ever dropped below my vsync frame rate, which is 60fps @ 1600x1200 with everything turned on. then again, i had a x800xt back when i played it...but didnt know that would make a difference from 1gig to 2gigs if you are vid card limited...

Quote:

---

Originally Posted by chinkgai

wow you noticed a difference with 1gig vs 2gig with a 6600gt in wc3:tft??

i recently stopped playing that game but before i stopped i remembered that game never ever dropped below my vsync frame rate, which is 60fps @ 1600x1200 with everything turned on. then again, i had a x800xt back when i played it...but didnt know that would make a difference from 1gig to 2gigs if you are vid card limited...

---

I like big maps and 12 group free for alls ;)

ooh i see, cant say i've ever tried that haahah

you should try aoe3 if you like wc

its my new addiction...i heard 1 gig to 2 gig makes a difference in this game as well...but i sold my two gigs already rrrggh..

Quote:

---

Originally Posted by sang

well,i digged 2 of my old screens for the trfc and trtw issue.With 2x256 Mushkin 3500 L2 those settings are piece of cake. I dont have anymore an A64 rig for testing but with your kindness we can find the exact difference.
Please,make another 2 tests for 3d2001. Find the highest memory frequency for your 2x256 sticks in order to be able to bench with the above timing parameters TRFC 12/TRTW 1 versus TRFC 14/TRTW 2.
Thanks in advance :toast:

---

TRFC=12 is not a problem... With my sticks, for anyhing above 260MHz I can't use TRTW=1 as it's BSODs all over the place, regardles voltage. I have had those sticks for a very long time and tried it many times. TRTW=1 would gain some performance, but TCCD can also be tweaked further (not TRTW=1, of course).

This is sort of off topic, but does the amount of ram not affect 3d01,03 or 05 scores?

Critical.Arrow