RV350 benchmarks ?

Uttar said:
....
Why did I post them, even though they're on a different setup?
Because otherwise, those scores wouldn't have made any type of sense.
And all the things I've given comparaisons for ( there are 2 or 3 others at nV News ) are things which are highly unlikely to be CPU limited.

But the score that was quoted here was Quake III, hence the wording of my question. :p Perhaps you could link the other comparisons?
 
It's not clear to me they are from specifically ATI, though they are from from "CeBit".

In any case, those score would more or less fall in line with about where I would expect them to for 3DMark and other tests that use shading to a significant extent. However, in today's games and with AA, I wouldn't be surprised to see the 9600 Pro outrun the 9500 Pro.
 
FYI, Uttar has a 3DM03 score of 3170. According to Hothardware an engineering sample of Mobility Radeon 9600 PRO running in a PC scores 3159 and that was running at 350/300.
 
I'm not understanding why the R9600 would be slower than the R9500. They're both 4x1 and 128 bit, but the R9600 should be clocked slightly higher? Also the R9500 has Hierarchical-Z disabled...

It makes sense for the R9500 Pro to be faster since it's 8x1, but I'm confused about the non-pro situation. In my opinion the scores just don't add up.
 
Oh, the non-PRO 9600! Remember, currently this is said to only have 200MHz memory, as oppose to 275 on 9500 IIRC.
 
Seems like there are rumours (and I mean that in the Inq sense) that ATi may make a few last minute changes... What though is puzzling -- surely it can't be a major architectural change, so it would have to be limited to core/clock frequencies, or am I missing something?

LW.
 
Yes, clock bumps, hence Dave's (the Wavey kind) vague noodlings about an extra power connector.
 
Not going to make a new topic but what do you think of that :?: :

Oh, and on the note of the 9800, the silicon is physically identical to the 9700, on a physical level all the chip does is run at a quicker operating frequency.

[....]

The 'source' wasn't an aticle, as I have already said it was a result of a briefing from ATI at CeBIT last week... Norbert Kuperjans, Technical Marketing manager for ATI Europe stated that the GPUs are architecturally identical that the silicon hasn't been changed

http://www.dearwandy.com/showthread.php?s=&threadid=356&perpage=15&pagenumber=2
 
RussSchultz said:
Maybe they cut some other parts out to make the die smaller?
According to the specs on ATI's site, the 9600 also only has 2 geometry engines, as opposed to the 9500's 4.
 
Most likely, Steve at DearWandy misheard or misunderstood what was said. If he truely believes that to be the case, he's more of a flake than I thought.
 
BRiT said:
Most likely, Steve at DearWandy misheard or misunderstood what was said. If he truely believes that to be the case, he's more of a flake than I thought.
That's what i think also, but well as i wasn't sure :?
 
In case you are wondering, Steve mentions an "edited down" thread in there.

That thread is the one where I quoted the discussion I had with him at Barrysworld (a thread he had deleted there, some of you may recall it...I have a copy of the replies I made on my hard drive since they needed to be split to be posted). What he stated exactly was that he "edited the remaining copy down so normal discussion can continue"...and what this process consisted of was excising the posts I made quoting him with no mention of them being removed left in the thread, and then the locking of that thread before any replies were added. This is the "PCZone GF FX Review" thread there, and if something seems a bit off about the remaining discussion, perhaps you'll understand what.

This is what he has claimed he will do with the comments he excised in this manner: "If I get any time in the next few weeks to address it, and feel it still relevant, I'll add my comments and re-merge it back into the main thread." This is the only reply to the questions and assertions I made that exists...in fact, after criticism started appearing, he stopped replying in the thread at all.

The thread you linked to is locked now as well, or there would be a mention of these factors from yours truly included.

I'm sure most aren't surprised, but does anyone else remain capable of being amazed at his conduct?
 
BRiT said:
Most likely, Steve at DearWandy misheard or misunderstood what was said.

Well that's what I'd like to clear up with him. He truly believes the 9800Pro is a con job by ATI (or he truly wants his readers to beleive it).
 
I don't think there is any con job by Ati. The 9800 and 9700 are nearly identical, but they are different chips with the 9800 having small fixes and tweaks. It seems perfectly resonable to me to increase the product ID by a small amount.
 
Even if the chips are physically identical, I don't see a "con job" in offering a higher-clocked versions (9800P, 9800) for the same price as last year's top end (9700P, 9700).

Does he consider the 4800/SE a "con job" as well? Or the GF4MX? :p
 
Back
Top