HAHAHA you mean the Charlies faked fanboi rants?Bell wrote:
You seen the reported yields on the GT300?
Last edited by GC_PaNzerFIN (2009-09-20 10:05:14)
He's not the only one coming away with it. I have often found Charlie does indeed exaggerate and blow things out, but lying about it not so much.GC_PaNzerFIN wrote:
HAHAHA you mean the Charlies faked fanboi rants?Bell wrote:
You seen the reported yields on the GT300?
I could link straight away to 100 anti-NVIDIA fanboi "news" from him that turned out to be complete bs.
lol you really think he is telling the truth? His whole site is paid by AMD. He left Inq because he was too biase even for them. You need over 100 silicon wafers to do yield calculations.Bell wrote:
He's not the only one coming away with it. I have often found Charlie does indeed exaggerate and blow things out, but lying about it not so much.GC_PaNzerFIN wrote:
HAHAHA you mean the Charlies faked fanboi rants?Bell wrote:
You seen the reported yields on the GT300?
I could link straight away to 100 anti-NVIDIA fanboi "news" from him that turned out to be complete bs.
Last edited by GC_PaNzerFIN (2009-09-20 10:14:54)
He's at ''semi accurate'' now I think (). He may have a point, the GT300 chips where always expected to have worse yields than GT200 at launch, and they where at like, what, 60%?GC_PaNzerFIN wrote:
lol you really think he is telling the truth? His whole site is paid by AMD. He left Inq because he was too biase even for them. You need over 100 silicon wafers to do yield calculations.Bell wrote:
He's not the only one coming away with it. I have often found Charlie does indeed exaggerate and blow things out, but lying about it not so much.GC_PaNzerFIN wrote:
HAHAHA you mean the Charlies faked fanboi rants?
I could link straight away to 100 anti-NVIDIA fanboi "news" from him that turned out to be complete bs.
He is saying GT300 yield is less than 2% at semi accurate.Bell wrote:
He's at ''semi accurate'' now I think (). He may have a point, the GT300 chips where always expected to have worse yields than GT200 at launch, and they where at like, what, 60%?GC_PaNzerFIN wrote:
lol you really think he is telling the truth? His whole site is paid by AMD. He left Inq because he was too biase even for them. You need over 100 silicon wafers to do yield calculations.Bell wrote:
He's not the only one coming away with it. I have often found Charlie does indeed exaggerate and blow things out, but lying about it not so much.
I think it was bright side of news that had a story back in june claiming 30% yields and I dont think even charlie was having a go that early.
Last edited by GC_PaNzerFIN (2009-09-20 10:20:06)
How many chips on a single silicon wafer are 100% functional from large quantity of wafers.CapnNismo wrote:
What do you guys mean by yields?
Last edited by GC_PaNzerFIN (2009-09-20 10:41:38)
No, it means that the ones that can't become GTX380s become GTX350s.CapnNismo wrote:
Gotcha. Does that mean that cheaper brands such as Sapphire or such get chips that are less functional than say, a card from ASUS?
eh you missunderstood it. They use the less functional chips in cheaper card models. All cards are manufactured at same factory, brands just put their own sticker on the card, add bundle and warranty and retail box.CapnNismo wrote:
Gotcha. Does that mean that cheaper brands such as Sapphire or such get chips that are less functional than say, a card from ASUS?
Exactly. If 16 of the 128 shaders of an 8800GTS 512 are broken, they make it into an 8800GT with the remaining 112 working shaders.CapnNismo wrote:
Ahhh, OK. So it's just a matter of throwing a perhaps less than functional chip on a card and then deactivating the bad parts? I guess?
Perfect, now I understand.Freezer7Pro wrote:
Exactly. If 16 of the 128 shaders of an 8800GTS 512 are broken, they make it into an 8800GT with the remaining 112 working shaders.CapnNismo wrote:
Ahhh, OK. So it's just a matter of throwing a perhaps less than functional chip on a card and then deactivating the bad parts? I guess?
Funny - when you compare the two pages perfectly side by side - nothing changes except the 5 in 50 becomes a 70. All that for an extra $100?GC_PaNzerFIN wrote:
Pre-order prices:
HD5870 price
HD5850 price
http://www.techpowerup.com/img/09-09-22/20a.jpg
there is a good reason for that, specs are released tomorrow so atm they have old HD 4xxx spec pages. The HD 5850 has less SP, and a lot lower clocks. Means over 20% slower than HD 5870. 5850 also has smaller PCB, cheaper components etc.CapnNismo wrote:
Funny - when you compare the two pages perfectly side by side - nothing changes except the 5 in 50 becomes a 70. All that for an extra $100?
Last edited by GC_PaNzerFIN (2009-09-22 07:25:02)
Last edited by CapnNismo (2009-09-22 08:05:42)
shader processorCapnNismo wrote:
What is SP? I've googled it, but I can't find it.
Last edited by GC_PaNzerFIN (2009-09-22 11:44:16)
Holy crap!GC_PaNzerFIN wrote:
edit: O LOL, Dell has reportedly reserved 90% of the first batch of HD5 series and only 10% will hit retail in first wave....
