ATI Mobility Radeon? HD 2600 vs. Nvidia Geforce 8600M GT 512MB

http://strony.aster.pl/pmpcfg/PC/ASUS_F8Sa/F8Sa_7.png - F8 with HD2600 score @ stock

 

thanks mastha212 for posting! But can anyone tell me what 3D marks really are (it shows the power of your graphics or something but I'd like to have a more complex answer)

 

http://en.wikipedia.org/wiki/3DMark

It's essentially a relative score of how powerful a gaming system you have, based on how well it plays through certain game snippets and does some rendering tasks.

 

thanks for the quick reply and link!

 

the ati hd2600 hasn't got a bad 3dmark 06 score...(notebookcheck and some posts above)

 

DDR2, right?

 

No idea

 

Hmm, from the comparisons with the 8600M GT I'd heard until now, it sounded like they were on par, but that 3dmark (thanks for that, BTW), is WAY lower than the 8600's. For those too lazy to click the link, the 2600 gets 3152 with the 1280x768 res. My 8600M GT DDR3 has 3870 with the 163.15 drivers, which is a BIG difference.

Either the drivers for the 2600 can be optimised a long way, 3dmark penalises the 2600 a LOT and in-game performance is actually much better, or the 2600 is not really all that comparable to an 8600M GT. Maybe the lesser 8600M GS model, but not comparable to the GT.

 

Thats what I have been saying the 2600 is the competitor to the gs while the xt is competitor to the gt. Some people disagree however.

 

Show me comparison in real games, not 3dmarks, and than we can start talking.

 

Or maybe the HD 2600 has two versions as well - DDR2 and GDDR3... is that news for you?

And the Asus F8sa has the DDR2 version (for these who wants to search, there are some chinese sites that say so...).

 

Most of 8600M GT card available are DDR2 versions (Vostro 1500, Inspirion 1520, Aspire 5920G) and these score some 3200 in 3D Mark 06. So there is no point in claiming those are incomparable to HD2600.
Te question is whether these HD2600 that we've seen so far are DDR2 or DDR3. As with 8600M GT most are probably DDR2 though.
The other thing is performance difference between those.
3D Mark clearly shows differences in points whereas we care about fps.

Take a look at the comparison test between Aspire 5920G and Asus G1S
http://www.notebookjournal.de/tests/262

I'm referring to average framerate- the difference in the highest fps achieved is far more visible which basically means the obvious- the fps is more core dependent.
The highest difference in favor of G1S is achieved in Colin McRae DiRT= 4fps. (G1S=22fps, 5920G=18fps). None of them is playable at these settings and both require some tweaks in game settings.
In other games the difference is either lower, nonexistent or even in favor of 5920G.
I know 600 3D Mark points looks far more impressive than 2fps difference in STALKER (with G1S having faster CPU) but still would you really say these are incomparable?
That's the same difference as between HD2600 and 8600M GT DDR3

 

Is this a poor, average or high score?
I'm not an expert at all

 

Thats pretty good.

 

Its more than 600. THose 3200 for the 8600 gt is 1280 x 1024 while the results for that 3dmark test were in 1280 x 768 res.

 

Nope these were done in 1280x800.
Dell Inspirion 1720 has 1920x1200 panel and has a 3D Mark score of 2930 achieved probably with default 3D Mark settings.

EDIT: Aspire 5920G C2D 2.0GHz, 2GB RAM, 8600GT 256MB DDR2, stock clocks: 3250 3D Mark 06 @1280x800
http://www.flickr.com/photo_zoom.gne?id=766316223&size=o
Thanks for @ bigspin for the test and screenshot.

 

Guys guys....read the entire thread before posting, page 11 has been totally off.

The HD2600 is comparable to the 8600GT, look at the last couple of pages. Its NOT comparable to the GDDR3 8600GT as of yet because nobody with a 8510p has managed to post any benchmarks but it is comparable to the 8600GT in say the Inspiron 1520, Sager NP2090, Compal IFL90, or Acer 5920G. Especially with Catalyst 7.8 which improved F.E.A.R from 26 to 30 FPS in the benchmark.

 

We know that, but people like Forerunner seem to insist on saying that it's comparable to the 8600M-GS.

 

C90s get around 3200 on 3dmark06 on 1280 x 1024.

 

Wow. Colour me surprised again then. Although I haven't been checking the 3dmarks of the DDR2 8600M GT's, I didn't think the difference would be 600+ points!

That F.E.A.R test is pretty good, pretty much on a par with what my G1S' 8600M GT manages, although it's not certain exactly which settings are turned on.

 

You know that's entirely unfair, as 3dmark06 has a significant CPU component

 

No no no. Ok the 3dmark score was 3152 in that pic with 1280 x 768 res with the 2600. Someone said that the 8600 gt gets around performs around the same around the same. I said no 512 mb 8600 gt gets 3000-3300 stock at 1280 x 1024 a higher resolution. So the 2600 is not that close to the 8600 gt.

 

Quoted for people in this thread who seem to have skipped my post entirely

 

And what CPU are you getting those 3000-3300 scores with? You cannot use the C90s scores as a basis of comparison, because 3dmark06 has a major CPU component in it's score, and a desktop CPU will add a significant number of points. The 256mb DDR2 8600M-GT with a T7500 scores about 2900 3dmark06 in 1280x1024. At 1280x800 the 256mb DDR2 8600M-GT with a T7300 scores about 3200-3400. So for the HD2600 to score 3150 with a T7300 (and unknown memory type, clock speeds, or drivers), I'd say it's pretty dang comparable to the DDR2 8600M-GT. 100-200 3dmark06 difference is not going to mean anything in real life gaming.

 

Actually cpu makes a difference yes, I don't argue there but not a major change. THis is debatable because different tests have been done with some people claiming that there is a small difference while others say there is a signifigant difference. This is a debateable point. Look at how you said that the ddr2 8600 with t7500 gets 2900, while I said 3000-3300? So how did I inflate the score? I said the base was 100 more than yours.

All I know that if Ackeron is right then I was wrong to a point. I just wanted to get stats before making the comparison and the 2600 is quite impressive so the xt must be very powerful. Regardless I still stand by my claim that the 2600 is still supposed to be the competitor to the gs, and all this means is that the 2600 blows away the 8600.

 

Because 3000-3300 is a major inflation. With a normal mobile processor at 1280x1024 and stock settings, the 256mb DDR2 8600M-GT will not break the 3k barrier, so to say 3000-3300 as the norm is a major misrepresentation. Look at the score breakdown on your 3dmark06 versus one on a 1520 or 1720, you'll see the difference in the CPU scores. It's a significant difference.

 

Lol. Read the last line of your other post. You said 100-200 points in 3dmark means nothing. Then why is my 3000-3300 statement such a big deal.

 

I honestly can't figure out what the argument is, if you read further back in the thread you would see what 3Dmark06 scores the HD2600 with 600mhz memory has gotten. Yet we can't determine anything because all we've seen so far in terms of real gaming performance is from an HD2600 with 400mhz memory, but those scores still were impressive at 30FPS F.E.A.R max. We have no idea if the HD2600XT or HD2600 GDDR3 (im pretty sure that different core thing was a rumor more likely the XT version is just higher rated silicon [see x1900xt vs x1900xtx]) will run F.E.A.R max @ 40+ FPS like the 8600GT GDDR3 in the Asus G1s can...

 

It means nothing in terms of real performance, but I just thought that when you imply that the "norm" for a card is 100-400 points higher than it actually is, it seems kind of misleading to those reading this thread who don't have as good a grasp on what theses scores actually mean as you or I do.

 

Max at what resolution?

 

it was all the same resolution if i recall ...1024x768 i think.

 

It's the same core, mobile version of the R600 (HD2900XT) From ATI's website, here's the spec.

Mobility 2600= 500mhz core/600mhz memory (GDDR3)
Mobility 2600XT= 700mhz core/700mhz memory

 

how about we all just be friends

and wait till someone decides they want to do a comparison or do benchmarks of their own..

 

lawl good idea... from what it looks like ati is going to win this one. the main reason is ddr3. most Nnidia geforce 8600m gts have ddr2

 

When I will get mine U can count on a small [or biger] review Im mostly interested in the graphics so loads of different game test should be included

 

good im looking forward to it

 

nice, try out world in conflict(demo) and call of duty 4 (beta). That'd be awesome to see how the Ati 2006 card performs with these 2 brand new games.

 

Hi!

I have a HP HDX with the Radeon HD2600XT. I have'nt read the entire thread but if i can run some benchmarks etc. Just tell me. I will do some testing this evening.

The HDX makes my 17" laptop look like a calculator. I love it

 

Do you have any modern games? If so, could you tell us what settings they run well on? Also, if you have a chance, could you run Everest and report back on what it says about your GPU?

 

There are several versions of Everest. Which one?

New games. Stalker, FEAR and Bioshock.

 

It would be nice to see how the FEAR performance test runs at Max/Max, and how Bioshock performs at native resolution (either a subjective analysis or some FPS data with FRAPS).

As for Everest, I was speaking of the newest free edition. I think it's 2.2 or something like that?

 

Could somebody please indulge me by answering some questions I'm sure have been done to death by now?

Which is faster: the 8600GT 512mb or the ATi 2600 256mb?

Thanks!

 

Which laptop with a HD2600 are you looking at? The 8600M-GT 512mb I assume is the DDR2 version, which will be inferior to the higher clocked (GDDR3?) version of the HD2600, though it will be on par or slightly better than the lower clocked (DDR2?) version of the HD2600.

 

I am considering a HP 8510p primarily- it has a DDR3 card right?

Thanks.

 

From what we know now, it appears to have an underclocked GDDR3 card, yes. It will perform slightly behind the GDDR3 8600M-GT, and slightly ahead of the DDR2 8600M-GT.

 

Thanks for answering my questions.

So would it be considered an upper or lower level notebook card? Could you run World of Warcraft and Counterstrike: Source at 1680x1050? How about Oblivion and soon to be released games like Crysis?

 

It's considered an upper performance level graphics card for notebooks, not the best of the best, but still pretty dang good (while still leaving you with enough battery life to be considered portable). It should have absolutely no problems with WoW and CS:S at high res, with Oblivion you won't be able to run on max, but still might be able to play smoothly at 1680x1050 (though not at high settings). Know one knows about upcoming games like crysis, though I'd say low to mid settings if you're playing at 1680x1050. For more specific info, look at how games perform on computers like the Asus G1s and the Macbook Pro.

 

OK cool. All I really want to do is play WoW and Age of Empires 3, I'll probably give the others a miss so it's great to know they will run at high settings.

Thanks.

 

I test out WoW and CS:S in my review here:

http://forum.notebookreview.com/showthread.php?t=169816

(long story short...both are fine )

 

Hey guys...Just got the Asus F3 series w/ the ATI 2600 in it w/ 2 gigs of ram and 2.0 ghz amd processor. I've used the driver modding tool to get the latest drivers for it and after I downloaded ati tool 0.27b2 to see what my core and memory clock speeds were so i could find out which version of the card i got.

The ati tool was giving me some goofy number of a 32.04 mhz core and memory. WTF is going on here? I've used ati tool on my desktop w/ no problems......any ideas?

I just want to know if i got the ddr2 or gddr3 version of the card.

ps (in the display settings it says I 512 dedicated vram on the card...didn't know there was even a 512 version)