ATI x1600 vs. nVidia geforce 7600

I absolutely have no idea which one performs better..
plz... help me..

 

I believe the x1600 is slightly better than the 7600, but I'm not sure exactly by how much.

 

Im interested in this too b/c I might either upgrade the the x1600 that AW is offering for my lappy or wait for the 7600.
Benchmarks anyone?

 

I'd get the x1600, but I'm slightly biased to ATi so my opinion may not mean much ^^;

 

The cards overall performance should be roughly the same. Maybe a 1 or 2 FPS difference either way and that would vary from game to game on which card gets the advantage. However, in games based more twoard OpenGL, the Nvidia card would have a small edge, and in DX9 based games, the ATi card would have the edge.

 

And more games are DX9 based than OpenGL are they not? The newer games anyway...

Please correct me if I'm wrong..

 

Here's an article describing the range of 3dmark 05 scores most cards get:
http://forum.notebookreview.com/showthread.php?t=46925

Here's an article describing how the video cards are set up: ie pipes, vertex units, etc.
http://forum.notebookreview.com/showthread.php?t=25266&page=4

both have excellent information.

 

Thanks for linking those KoOni.

 

but i seem to me that mobility x1600 outperform GO 7800GS though...

 

X1600 performs the same as a 6800 Ultra

 

The x1600 gets thoughrilly beaten by the 6800 and more by the 6800U. The x1600 only has four ROPs while the 6800 has 12. PLUS the 6800U has over twice the memory bandwidth.

PLEASE check my graphics charts before asking these questions.

 

the X1600 does perform the same as 6800 Ultra. sometimes better. Check benchmarks from various websites

hey i noticed ur laptop is the same as mine ^_^ . except i got an X700 256mb and DDR2. 1694 is great value for performance isn't it?

 

You mean like this?

http://www.anandtech.com/video/showdoc.aspx?i=2556&p=7

and this?

http://www.anandtech.com/video/showdoc.aspx?i=2556&p=6

 

FWIW, the X1600 has 12 pipes compared to 8 with the 7600. I suspect 90+% of us really couldn't tell the difference between the performance.

 

Meaker, you are comparing two different beasts. The desktop x1600 does get beat pretty badly by a 6800 ultra and 6800 gt, but in the laptop world it is different. a X1600 performs a little bit better than the 6800go and the 6800 go ultra performs a little bit better than the x1600. At least that is what reviews seem to show!

 

I think you will find that the 6800U go beats a 6800GT in most cases and even sometimes the desktop 6800U (400/550 version) due to a very high core clock. As such the x1600 cannot keep up, links to these reviews would be interesting as so far I am the only one who has done so. I know that the 6800go non ultra is quite a step down but I still believe the x1600 in most cases still is too far cut down from the x1600xt in that review due to a much lower clocks. (470/470 vs 590/690).

http://www.anandtech.com/video/showdoc.aspx?i=2356&p=1

So if anything can be taken from the desktop review I posted is the gap should be LARGER for the mobile versions.

 

The Go6800U will outpace the X1600 by a good bit, I agree. Higher memory bandwidth, faster clocks, and more pipes = faster card. The X1600 was never designed to compete with a Go6800U in the first place, so it's not a fair comparison.

The X1600 should be a bit faster than the Go7600 based on paper, but you can't go wrong with either.

Chaz

 

I will say it again, desktop gpu's do not compare to laptop gpu's.

Here is a good comparison between the X1600 and a 6800go.
http://www.laptoplogic.com/reviews/detail.php?id=101&part=full&page=8

Here are the scores
-------------6800 go------------x1600 mobility---
3dMark01: --17654----------------18149------
3dMark03:---7840-----------------7167------
3dMark05:---2256-----------------3810-------

Some things for this review. The x1600 laptop has a faster cpu but it is only using the 128mb x1600 instead of the x1600 with 256mb (which the w3j has as well as having 512mb hyper memory)

Needless to say the 6800 go is comparably on the old benchmarks but the new ones the x1600 rocked it.

 

Well the (normal) x1600 is actually a tiny bit slower than an 7600.
If someone knows german go checkout www.notebookcheck.com and especially their reviews of the Toshiba A100-153 and the Acer Aspire 5652.

Same CPU - Core Duo 2300
Same MOBO -Intel 945PM Express Chipset
RAM - Acer slightly faster @ 667 mhz Toshiba @ 533 (maybe this makes the difference?)

GPU
Toshiba: ATI Mobility Radeon® X1600 512 MB HyperMemory (256 DDR VRAM+ 256 System RAM)
Acer:GeForce Go 7600, 256 MB

The Benches: (Acer/Toshiba)

3DMark 01: 16968 / 16251
3DMark 03: 7877 / 6132
3DMark 05: 3428 / 3400
3DMark 06: 2108 / 1766

PCMark 04: 5008 / 4411
PCMark 05: 3710 / 3766

DOOM 3(Acer/Toshiba):

Details: Low 82,8 fps/80,8 fps
Details: Medium 75,0 fps/72,3 fps
Details: High 56,8 fps/53,0 fps
Details: Ultra 37,1 fps/37,2 fps

You cant realy say who has the edge here,the results are so close that its almost sure its the Acer's faster RAM that makes the difference.

 

JackBauer is correct, the chart in the link KoOni gave was wrong. the 7600 does not have 12 pixel pipelines, the x1600 does. I think for Linux users out there, nVidia is the way to go. Otherwise, I'd say it's a toss up overall, and performance overall should be comparable.

 

The Go6800 benches a lot higher than that in 3DMark05.

Remember, they're only 3DMarks . . not real games, so you can't rely on them for accurate data. Not that I'm nitpicking your efforts, but basing them on more solid ground would be better.

The X1600 is more advanced than the Go6800, so I can see that it would pull ahead in some newer games.

Chaz

 

benchmarks, are biased. some tests need a better CPU, some rely on a better GPU, some like more ram, and they will raise your score.

barrok--
were those two laptops identical other then the GPU?

and as chaz said, just becuase one setup benches higher then another, does not mean it will perform better in games, and vice versa.

pb,out.

 

I stated in my post that they were almost identical except the x1600 had a little faster CPU I believe. Everyone always says not to listen to benchmarks but benchmarks are how people decide what to buy. You look at benchmarks for cars (speed, gas) you look at them for laptops, heck you even look at them for Graduate School and what field you want to go into. Yes, some have different factors but you can't just discount the benchmarks you don't like the outcome of

 

Ouch the 6800go is having issues there as those benchmarks are telling me its only 25% faster than my x600 which is bull.

The 6800go should be getting over 4000 in 3dmark 05 and nearer 10k in 3dmark 03.

Since I am a futuremark.com forum member of over 8000 posts using bias/sqewed 3dmark results is not going to work

 

http://www.powernotebooks.com/

go to the PowerPro A 15:4

you mean to tell me..that an x700 (2595) with 128mb ram will outscore a 6800 (2256)?

those benchmarks are wrong. there is no way possible that an x700 scores higher then a 6800..this also somewhat proves that some benchmarks are biased towards certain things.

pb,out.

 

The other thing you have to take into consideration, that most gamer's do is OC. I mean my card did like 3200 or something in 3dMark05 and then I OC'ed it and get 4573 and that isn't the highest I can go.

Then again...it will also depend on the manufacturer, because my Dell was way underclocked.

 

If you have such futuremark expertise, then you would know how to use the ORB instead of posting links to desktop cpus. Looking at the Orb, the fastest mobility x1600 is as fast as the two listed 6800 ULTRA go's in 3dmark05 (the drive approved ones). Most likely those x1600's are oc'd but those are the only results we have. Also, there is no listing on the orb for basic 6800 Go's, but judging from the 3dmark05 performance of the 6800 Ultra Go, I can see how the 6800 could get low 3dmark05 scores.

 

Here is another benchmark you can go look at

http://www.notebookforums.com/portal/modules.php?name=Reviews&file=viewreview&id=8725

With a 6800 Go:
STOCK CLOCK RESULTS (290/590):

3DMark03: 8065

3DMark05: 3322

OVERCLOCKING RESULTS (401/685):

3dmark03: 10119

3dmark05: 4337

SO yes, Oc'd by a good load the 6800 Go gets over 10k in 3dmark03.

Even oc'd it loses to the x1600 mobility in 3dmark05. It's oc'd results are close to some stock results for the x1600 in 05.

 

Benchmarks don't have to be biased to provide results that don't conform to most peoples results. Cpu, Ram, drivers, spyware can all lead degredation in performance rather than a "conspiracy" to belittle a product.

 

they can also be using different components..more/less ram, higher/lower CPU speeds (overclocked or not), higher clocks on the card (overclocked or not).

these "benchmarks" should be done on a fresh boot, to ensure best results compared to a system that has been running for months with other things installed on it.

BUT the first benchmark you provided the link for, was WAY to low. it was lower then an x700..which just isnt accurate.

pb,out.

 

I read in a review overseas for new asus A8Jm which has 7600go benches were as follows 3791 3D 05 and 2280 3d 06 with stock ****py drivers they all come with

 

You will see that the 4000 score for a 3.2ghz p4 and 6800go @ 324/350 (nvidia says 6800 should be around 300/300) is about stock as is the story for the 4000 scores for the x1600 (the 5000 score is obviously more overclocked and tweaked as 1000 points (25%) does not come from no where) so in 3dmark the x1600 is about equal to the 6800go then. However the x1600 does better in 3dmark than it does in most real world (games) benchmarks therefore the place on my graphics charts will always be 6800 (non ultra) followed by the x1600.

Now considering the 6800U go has a 50% core and a 92% increase in mem bandwidth over the vanilla, the x1600 is never going to get close.

 

What is your link to this 6800 go. There is no 6800 go that I can search by on the orb, only 6800 Ultra go.

 

http://service.futuremark.com/compare?3dm05=1545417

Right there.

 

http://www.extremetech.com/article2/0,1697,1714980,00.asp

There is more tests showing the weaksauceness of the 6800 go.

 

First of all, that's a very old review. Second, isn't that the DDR1 version of the Go6800? The true version uses GDDR3.

Chaz

 

The true version? Age of the review doesn't matter much in laptops. The cpu's aren't going to vary a ton. Do you guys just not like the results or something? X1600 clearly beats the 6800go in many instances. The 6800 go wasn't the highend remember, the 6800 ultra was. So ATI's mid range should beat the old midrange 6800, that should make sense? why is that hard to believe? The 7600 mid-range and x1600 have a good battle and the 7600 is the replacement for the 6800. Sheesh!!!!

 

here are the specs off nvidias website

GeForce Go 6800
Memory Interface 256-bit
Memory Bandwidth (GBps) Up to 35.2
Fill Rate (pixels/sec.) 3.6 billion
Vertices/Second 470 million
Memory Data Rate 700 MHz DDR/DDR2
1100 MHz GDDR3
RAMDACs 400 MHz

Since i haven't followed it until recently, it seems that they switched the ram type they used during production? Not sure Also, I guess I am wrong that the 6800 go was replaced by the 7600. I guess the 7600 replaces the 6600 and the 7800 go replaces 6800 go, my bad

 

The original Go6800 used DDR1 - once they switched to GDDR3, which was shortly after if I remember correctly, there was a big boost in speed.

Barrok - I have no problem with your arguements, and nor am I arguing against you. All I'm saying is that the Go6800 is more powerful than it is made out to be here.

Chaz

 

So then, alot of these benchmarks with the 6800 go getting rocked must be from ddr1 days. Makes sense. Thanks for explaining that! I still believe the x1600 is faster in most things from the benchmarks I have seen, but by no means is the 6800 go a terrible card. It seems pretty equal except that the x1600 is newer! Anyway, both cards are great. Wish we had more reviews of the 7600 to compare!

 

So do I - Nvidia doesn't seem to be getting the Go7600 into very many notebooks. There's an Acer or two in the U.K. that have it, an Asus too if I recall, can't think of anymore.

Perhaps we'd be able to derive a rating system from comparing the desktop cards . . but there's variables involved there too.

I still think that whether you get the X1600 or the Go7600, they'll both be superb cards for the midrange.

Chaz