New Radeon core clocking at 450 Mhz and burning 11 watts

ahhh good, for a second there, i thought this was about the upcoming '420

I'd like to see 750 Mhz * 16 pipes for the Xbox derivative of R500.
 
Ati knows how to build a cool chip . I wonder if ms didn't go with ati for more reasons than just the new xbox chip . Perhaps they wanted an xbox that could some day be portable ?
 
Or simply just a cooler (temperature wise) design. Lower temps will put less strain on the hardware, which will help the components last longer.
 
DeathKnight said:
Or simply just a cooler (temperature wise) design. Lower temps will put less strain on the hardware, which will help the components last longer.
also will let them design a compact xbox 2 .

But i like to look further :)
 
Deadmeat said:
Hopely, XGPU2 core hits 700 Mhz when migrating to 90 nm???

I thought you said in another thread .09 tech only made the chip smaller, not cooler or faster? :LOL: Or does that only apply to Sony's chips? :rolleyes:

Anyway, xgpu will most likely not "migrate" to .09, I'd think it will be designed for that size from the start, or possibly something even smaller.
 
I belief with ATi behind them, Xbox2 SHOULD have THE BEST IQ again. MayB along with GC2. Cant wait to see the FSAA and the AF and the Mipmapping of X2!!! I think the texturing and shadering should be good.

Funny thing is, good lord to PC hardware devs for consoles. This gen we have PVR, ATi and Nvidia, and them do contribute to giving many somewhat PC "quality" IQ on good TV. IQ is VERY important ImhO to reach that CG/Photoreal/Toystory graphiX. :)
 
Guden Oden said:
Deadmeat said:
Hopely, XGPU2 core hits 700 Mhz when migrating to 90 nm???

I thought you said in another thread .09 tech only made the chip smaller, not cooler or faster? :LOL: Or does that only apply to Sony's chips? :rolleyes:


Yeah, it only applies to Sony's ones, cause we all know Sony uses plastic transistors... They use Lego! They just melt as soon as you turn them on... ;)
 
Guden Oden said:
Deadmeat said:
Hopely, XGPU2 core hits 700 Mhz when migrating to 90 nm???

I thought you said in another thread .09 tech only made the chip smaller, not cooler or faster? :LOL: Or does that only apply to Sony's chips? :rolleyes:

And nevermind that he's quoting the thermal output of a bastardized IC for the portable marketplace. Lets forget for a moment that the 9800TX board will draw ~70-75watts before Overdrive.

And, if I wanted to be like Deadmeat (which is scary in itself) and forgo any rationality, sound basis, or context, I could post what Eric stated:

[url=http://www.3dcenter.org/artikel/2003/11-06_english.php said:
ATI's Eric Demers[/url]]In general, with shrinking processes (150nm, 130nm and now 90nm), active power has gone down per transistor (smaller transistors and voltage lead to less power outputted when switching), but our transistor counts have increased at a larger rate, so that the active power has kept on going up. But with 130nm and even more with 90nm, static power is getting higher and higher. In a normal CMOS process, one can think that if a transistor is not switching, it's not using power. That's roughly right, but there is something called "leakage", which is a small current that leaks out of a transistor, even when not switching. As dimensions have shrunk, leakage currents have gone up by about an order of magnitude each time. This is leading to the problem that non-switching power is getting bigger and bigger. With 90nm, it's going to more than ½ the total power. This power is directly proportional to the number of active transistors, or plain die size. Consequently, as we shrink down, active power sort of goes down, but passive power goes up. With 90nm, power consumption is going to be a "Big Deal", even when the chip isn't doing anything.

Having said all that, even the high-end desktop parts cannot keep on increasing their power consumption indefinitely.

So teh ATI has admitted that their 90nm chips can't use more power on even a desktop and since the power intake is becoming bounded proportionally to the absolute number of transistors (due to leakage), there's no reason to go beyond 130nm to 90nm where the IC will consume 1/2 it's power just sitting there. So, basically, ATI just admitted that CMOS scaling has ended and that their XGPU2 will basically be a 130nm part that can't surpass the PC due to power/thermal requirements, which we'll base off the 9800XT...

:rolleyes:^10 ;)
 
Vince said:
nevermind that he's quoting the thermal output of a bastardized IC for the portable marketplace
There is nothing 'bastardized' about it. The core was architected to suit the requirements of both markets from day one.
 
Dio said:
Vince said:
nevermind that he's quoting the thermal output of a bastardized IC for the portable marketplace
There is nothing 'bastardized' about it. The core was architected to suit the requirements of both markets from day one.

Ok, so are we just going to play PR games now? "Bastardize" is defined as "To lower in quality or character; debase." The mobility Radeon is definitely of lower character and stature than a preformance IC such as the NV2A, or a contemporary ATI analog such as the 9800XT.
 
I'm slightly confused. You said 'a bastardised IC'. The IC in question was designed for the market that it hit; it is the best performing chip with the best image quality in its market; how does this fit in with your definition?
 
Dio said:
I'm slightly confused. You said 'a bastardised IC'. The IC in question was designed for the market that it hit; it is the best performing chip with the best image quality in its market; how does this fit in with your definition?

Yes, you are. The ultimate object of quoting the power requirements of this IC [Mobility 9700] are to relate them to what can/will be done on the XGPU2 in his mind. As Deadmeat most clearly stated:

Deadmeat said:
New Radeon core clocking at 450 Mhz and burning 11 watts... Hopely, XGPU2 core hits 700 Mhz when migrating to 90 nm???

So, he's comparing a mobile IC to a high-preformance IC. I think we can agree that each niche has a distinct equilibrium of power requirements, thermal output, preformance and price - do you disagree?

If you agree to the above, you can do the approximate math and see that a mobile IC will have a lower equilibrium point concerning preformance as related to, say, power requirments - which is why it has a vastly lower shading rate, pixel rate, logic count, clock speed, etc.

Therefore, we can conclude that a mobile IC is lower in character than a High-Preformance IC akin to a contemporary NV2A. It is thus a bastardized IC in comparason. Why must I explain stuff like this?

This clearly stems from where you work and is akin to nothing more than PR spin which must allways put something in the most positive light, just because. Everyone else is questioning why we're comparing the miles per gallon of a hybrid car to a Hummer since they're not related, and low and behold there's the hybrid car PR man defending it's off-road preformance relative to something not even near the Hummer's class. But... But... it's the best at going up parking ramps!!!
 
Why must I explain stuff like this?
I would suggest that saying "Sorry, I didn't mean to say that Mobilitiy 9700 was bastardised, I meant to say that it's not relevant to his argument" would be perfectly acceptable.

This clearly stems from where you work and is akin to nothing more than PR spin which must allways put something in the most positive light, just because.
I obviously can't talk about future ATI technology, so I don't. I do make statements about current ATI technology when people make misstatements about them. I think most other people on this board will be aware of this.
 
Dio,

I don't see why you object to calling the 9700 mobile "bastardized", when clearly it is a cut-down version. It's missing half the pixel pipes, half the vertex shaders I believe and the hyper-Z functionality too.

Do you have any actual arguments beyond the phrasing of Vince's post? Because unless you do, why keep replying unless it's just for the sake of arguing?
 
Guden Oden said:
I don't see why you object to calling the 9700 mobile "bastardized", when clearly it is a cut-down version. It's missing half the pixel pipes, half the vertex shaders I believe and the hyper-Z functionality too.
It's a stronger argument for the term, but still wrong. The original statement was that it was "a bastardised IC for the portable marketplace" - which it was not, as that particular chip was designed for both portable and desktop marketplaces. It is what it was intended to be from day one (day one of the design of the R300 family, too - they were not really designed sequentially) - so it cannot be 'lowered in quality or debased' because it would have to be lowered from somewhere.

Do you have any actual arguments beyond the phrasing of Vince's post? Because unless you do, why keep replying unless it's just for the sake of arguing?
Not really. I was just attempting to stop him applying his spin - which may be to counter someone else's spin, but I have no comment on the other person's spin because he stated only correct facts about current products and his spin was to extrapolate to future products. The spin I objected to was based on a distortion of the facts which I was asking him to withdraw. I agree that it's a largely pointless argument, I'd just prefer people to get their facts straight.
 
Dio said:
Not really. I was just attempting to stop him applying his spin - which may be to counter someone else's spin, but I have no comment on the other person's spin because he stated only correct facts about current products and his spin was to extrapolate to future products. The spin I objected to was based on a distortion of the facts which I was asking him to withdraw. I agree that it's a largely pointless argument, I'd just prefer people to get their facts straight.

There are no "facts" which you're challenging; merely chasing mythical "spin" which didn't exist in the use of that particular phrase. In case you still don't get it, it doesn't matter how the IC was devised with what rational behind it; the fact remains that it's a cut down/bastardized IC when compared with the high-end desktop/console ICs. Again, it could be the best damn hybrid car, which is in leagues of it's own when compared to that niche - but it's still a bastardized/poor-man's Hummer if you so choose to compare them, which DM did. I have described comparable parts from other IHVs in similar ways. It's correct semantically, it's merely that you don't like it. Deal with it.

If you really cared about the "facts" and the proper utilization of knowledge, which you don't, then you would have replied to DM and pointed him towards the power consumption by your higher-end ICs fabed with a low-K dielectric.. Something which one would assume is much more analogous to any future IC that would wind up in the XBox2. But, that wouldn't exact make the product look as good, now would it? Especially in light of the comment on CMOS slowing.
 
Back
Top