FWIW X1300 benchmark scores, maybe?

Acert93 said:
:oops:

Let me check my pants... egads! Obviously I would like to see some official stuff, but screenshot of the chip and 3DMark05 picks go a bit to showing this is legit.

X1300 beating a 6600? The GT is in the 29xx range. Your talking about an X1300, a low end part, beating a high-midrange part from two years ago (9800Pro, ~23xx) and last years low-midrange (6600, ~25xx) and very close to the best midrange product (6600gt, ~29xx).

This is really good news for the "baseline" performance. The R9200, FX5200, GF6200, X300 (and X600) are just dogs. A really bad balance of (non-)performance and features.

Looks like the X1300 can play games like Doom 3, Half-Life 2, FarCry, BattleField 2, etc... at decent resolutions and most of the features at respectible levels. This is great news for gamers, and great news for ATI. Obviously nice to have the fastest flagship model, but many more consumers will have low end products comparatively. And in the long run the hardware the low end guys has determines what features the top end guys get. If they can get this out the door for $100 (which sounds reasonable because a 6600GT can be had for $140) then this is indeed exciting times for consumers.

I wonder how much of the R520's architecture is shared with R515?
Click to expand...

but in fact the 3dmark05 result of 6600gt is 36xx+ when use a A64 FX55 and the FW 7X.XX.
 
cho said:
but in fact the 3dmark05 result of 6600gt is 36xx+ when use a A64 FX55 and the FW 7X.XX.
Click to expand...
That is why we need some real reviews with the same setup. :!: In this case we are comparing P4/Athlon setups with different memory configurations, etc... so nothing is conclusive (thus all the ~, xx, and ifs!)

Still, if a X1300LE is in the ~28xx range that is very impressive. That means the low-end products are moving up the performance ladder which is good news for the market and gamers. I cannot remember the last time that a low end part was actually relevant for gaming. :???: Heck, at these prices we may be able to get Jawed to upgrade to this or NV's competing part. I mean someday he needs to get off that Radeon 7000! For $100 or so this kind of performance is a steal.
 
Interesting that dispite the extra complexity it still seems smaller than the x600 (taking into account process shrink I was expecting it to maybe be bigger). From what I can tell it's the same height, but about 7% thinner. Thats pretty impressive really, twice the performance (x1300 pro?), smaller chip...
 
cho said:
but in fact the 3dmark05 result of 6600gt is 36xx+ when use a A64 FX55 and the FW 7X.XX.
Click to expand...
HKEPC X1300LE results are from P4 XE, so comparison to 6600GT scores from an Intel platform might prove more useful - as per Acert93...
 
Last edited by a moderator:
Screenshots in Age Of Empires III ?...with HDR

The videocard is the x1300 LE ?, 450 mhz core clock /400 mhz memory ? - 2.5ns memory

rv5156ys.jpg


http://chilehardware.com/modules.php?op=modload&name=PNphpBB2&file=viewtopic&t=9308


 
Last edited by a moderator:
Galduta said:
Screenshots in Age Of Empires III ?...with HDR
Click to expand...
HDR is expected of ATI's next batch of cards. But note the lack of AA. Doesn't necessarily mean that ATI doesn't support MSAA with FP16 rendertargets, but I'd say it seems teeny tiny bit less likely than before seeing those shots (though obviously they could simply be shots on an NV4x with HDR enabled).
 
Chalnoth said:
HDR is expected of ATI's next batch of cards. But note the lack of AA. Doesn't necessarily mean that ATI doesn't support MSAA with FP16 rendertargets, but I'd say it seems teeny tiny bit less likely than before seeing those shots (though obviously they could simply be shots on an NV4x with HDR enabled).
Click to expand...


Well it's supposed to be the X1300LE and at 8fps it would be pretty silly to turn on the AA. This may not be the only reason why AA isn't on, but I guess it's understandable for an entry level card to have such a low frame rate in such a game.
 
In the second image,
top left corner,
the hill, just above the fence,
there is some weird texturing going on there. diagonal lines running through it. Although this does match up with the shadow direction, and badly done shadow mapping usually gets these sorts of artefacts, yet it continues down the hill and only occurs on grazing angles to the view... odd, no?
And the black edge on the left side of the screen is odd too.

Question is, what sort of performance would you get on a 6600 running that resolution/scene complexity...?
 
That’s a pretty interesting score for a X1300LE.

If it’s a 4 pipe card clocked 450/400. Compare it to an 8 pipe X700pro clocked 425/430. The X700pro scores maybe slightly higher in 3Dmark2003 and 3Dmark2005 at very similar clock and memory speeds. Same 128bit memory bus.

That would mean the pipelines on the R515 are ~ 2x as efficient as on the R400’s. It would mean a 16pipe R520 might have similar (or close) performance to a 32 pipe R420 clock for clock.
 
Blastman said:
That’s a pretty interesting score for a X1300LE.

If it’s a 4 pipe card clocked 450/400. Compare it to an 8 pipe X700pro clocked 425/430. The X700pro scores maybe slightly higher in 3Dmark2003 and 3Dmark2005 at very similar clock and memory speeds. Same 128bit memory bus.

That would mean the pipelines on the R515 are ~ 2x as efficient as on the R400’s. It would mean a 16pipe R520 might have similar (or close) performance to a 32 pipe R420 clock for clock.
Click to expand...

Yes, my exact thinking too...I only hope those benchmarks are true, because if they are, that X1800XT will truly be a beast!
 
You must log in or register to reply here.

Similar threads

N
fraps Benchmarking
2 3
Replies
45
Views
10K
WaltC
W
Back
Top