Right, because you obviously know more than nVidia does on the launch of their product. :roll:
No, but I obviously know more than YOU which is what we're talking about here. You're saying nV has/had a set date for September when they hadn't even been able to test their design yet. So now YOU are saying that you know more than nV, because even nV didn't know what their fate was until they tested their part, so the assumptions you make (which are a giant leap from anything Geo would support you ascribing to him I'm sure) are the same as the people who 'knew' at the beginning of the year that these parts would be here this summer right after Vista launched, because both had due dates of the summer. Sound familiar? :roll:
Woah, wait a minute. Just a day ago you were saying - "The new cards are a MINIMUM 3+ months away."
Hmm, let's see. 3 months from the end of August (x1950XTX launch) is the end of November - not October? So which is it? 3+ months minimum or October? :?
Yes they are a minimum of 3+ months Away. 3 Months from today is the End of october, I didn't say 3 months after the launch of the X1950 next month. Perhaps if you were better at reading what people are saying instead of making up your own arguments, then you'd have less difficulty understanding what they're saying. Funny how you jump on thing yet can't focus on that one thing you jumped on. You wanna play semantics, then learn to actually quote the correct statement. There's a full stop period there, not relating that statement to anything else. Thus END OF OCTOBER, understand or do you need me to bring out a calander for you. This is why eVGA is still not a safe step-up option, simply because it wuld be outside the 90days, if they hit the late October target instead of the spring's optimistic Sept target.
First off, Radeon x1950XTX is set to launch, according to ATI, between Aug. 23-27. Three months after would be Nov. 23-27 - sorry to bust your bubble, but Nov. 15 is the middle of November.
Sorry to bust your bubble by folowing you down your own made up timelines, but mid November be it 1 week different or not is still outseid of my 3months statement, and really, anyone worrying about the X1950 versus waiting is pretty foolish to worry solely about products without launch dates, especially since this one will be far cheaper thanthe competition and something they can use, while the G80 is still a paper nothing without a launch date, so what's your point.
Actually the R520 did launch on Oct 5th, your lack of understanding of that show your myopia. Or was there another chip that launched instead. Perhaps some focus on your part would help.
The Radeon x1800XT wasn't available until the first of November, despite being announced on Oct. 5. So I guess I consider "launch" as being available, whereas you must think it just means a mere announcement.
No actually it's your lack of knowledge of this industry despite your statement in the other thread of knowing so much more than me...
The R520, in the form of the X1800XL launched on Oct5 to wide availability, the same R520 with higher clocks and faster memory and bigger HSF was launched again as an X1800XT model. So the R520 did indeed launch on Oct5th, unless you have evidence otherwise. SO I guess we can put that one to rest with the knowledge that you were just plain wrong. No point in beating a dead horse, so, since you can't get your past right why even bother with your predictions of the future? :roll:
So I can see how our difference in opinions would make this statement controversal (but I think more people would agree with me that a launch is when there's availability
).
Oh, but that's what I said now isn't it, the
R520 WAS available on OCt5th, and I think everyone would agree that I was right and you were wrong, because once again you DIDN'T FAQin' READ what was said, and this time YOU chose the words. YOU said R520, not X1800XT, see I can read, perhaps you should get someone to help you.
Geforce 7950GX2 hard-launched and has been in good availability ever since.
GF7900GX2 didn't hard launch, and considering the availability of the R520 before launch just like some of nV's parts I guess that makes you doubly wrong.
Lol, so I guess you'll be happy as long as it works (that's some mighty strong ATI loyality you have
)
Yep, that's usually all I look for in a computer part, that it works and is good bang for the buck, I don't look for salvation in my computer hardware/companies, unlike you obviously. :roll:
Well, for one, the arbitur units that tell a shader what to "be" in that moment can cause a bottleneck, with workloads literally waiting for the shaders to decide what to do.
Except being parrallel in nature and not assigned to a 'pipeline' anymore that delay is inconsequential, and will be nowhere near the delay caused by the ROPs.
Secondly, you have the fact that a specialized shader will always be better fit than a generalized one, until the workloads in a game, between pixel and vertex work, can actually commensurate to the point where a USA would be advantageous. That's not the case, as you should obviously know.
Yes, I do know but unlike you I see that as an added bonus to be added later and nearly irrelevant at the start which is why the G80 and R600 will be very close at the start, yet you see it as a dissadvanage which is far less proven, and speaks to your FUD mentality, the operationality of the unified design with unoptimized development had been shown with the X360 all of whose early games, have minimal optimization for it's unified design, one of the few titles to truely take advanatage has been Oblivion.
The benifits of unification have already been proven with emulation in DX9 hardware, once it's not emulated and done strictly in hardware then it's going to be faster still. Either way you're argument bears little weight on what you say, just FUd without even a logical backup.
Just like with R520, right?
Well if that's your only defence to me saying you don't know jack, then let me end our discussion with , G80...just like the fx?
Well, first off, I actually was very critical of a June launch, noting it's impracticality.
I don't care man, scepticism is not what you're holding now. I have a feeling that both nV and ATi are going to bust their humps to get these new cards to market in time for people to put them on Xmas wish lists, but I'm not naive enough to believe a date picked in the spring (which you still haven't shown as being anything more than a guess).
And second off, fair enough about you believing September. But to throw out some 3+ figure (which you know have seemed to change) was quite futile.
Futile, nah, based on first revision silicon nV will be pushing very hard to get it within the month of september, more likely the month of October, and IMO, it's be the end of October, with ATi hiting the end of Novemeber, thus 3+ months from now. Your arguing aginst that with little more than your assumptions based on statements probably formulated before there was a tape-out let alone working silicon shows you've got nothing worthy to counter point.
Perhaps you should show Geo's comments, because I'd love to see the date in September he picked and not the Q3 statements Huang was making, and had been since the March conference calls.
C'mon give me an exact date that puts them on track for September something more recent, something that doesn't just refer to that early date and says they need to rush to meet it.
Or once again did you misread someone else's statements when using them as your tools for an argument? :roll: