POPPUR爱换
标题: GF104/GTX460 has a huge die [打印本页]
作者: 我有3个马甲 时间: 2010-7-22 10:04
标题: GF104/GTX460 has a huge die
http://www.semiaccurate.com/2010/07/21/gf104gtx460-has-huge-die/
GF104/GTX460 has a huge dieEconomics of a failed architectureby Charlie Demerjian July 21, 2010
WHEN WE SAID that Nvidia's Fermi architecture was wrong, most people didn't understand what we meant, they focused on the chips. With the release of the GTX460, one look at the die size puts the problem clearly into focus, and it is quite ugly for Nvidia.If you recall, we were the first ones to say that the GF100 was huge, 529mm^2, months before anyone else had hard data. This was backed up 8 months later by some pictures on 【请勿利用PCINLIFE资源打广告，谢谢合作】. We can now tell you about GF104 die, it is 24.9mm x 14.7mm, and the die size is 367mm^2. (Note: The dimensions are rounded to one significant digit after the decimal point, they are a hair larger so multiplying those rounded numbers gets you 366mm^2. This is not an error, the 367 number is correct, some mad scientists directly measured the die.) For some odd reason, Nvidia didn't give out any die size numbers when talking to the press, almost like they didn't want people to know something before the reviews went up.
The closest site to publish an accurate number was the TechReport with 320mm^2. For that, Scott Wasson wins the mostly-not-crushed-much individually wrapped Kit-Kat stick that wound up in my travel bag after Computex. Congratulations Mr. Wasson, see me at IDF for your prize. No one said that the GF104 was larger than it's closest rival, AMD's Cypress/HD58xx, and that is what Nvidia didn't want you to know.
The problem is simple for Nvidia, the economics of this part don't work out, the underlying architecture is wrong, so the resultant parts start out with an uphill battle. This is Pro-Ablem for Nvidia, not for the end user. If the GTX460 is priced at a loss, the consumer shouldn't care, they get a deal, and that is the end of it. Retail buyers rarely care if the part is making Pro-Afit for the manufacturer.
Cypress/HD58xx is 334mm^2, which makes GF104 about 10% larger. If yields are the same for both parts, that would mean GF104 costs 10% or so more than Cypress to make. Sources in the fab community tell SemiAccurate that the GF104 is nowhere near the yield of Cypress.
As a humorous aside, both chips are made on the same process, TSMC's 40nm, and literally at the same fab. AMD managed to cram 2.15 billion transistors into 334mm^2, about 6.44 million transistors per mm^2. GF104 has 1.95 billion transistors in 367mm^2, about 5.31 million transistors per mm^2. This means AMD's Evergreen architecture is over 20% more space efficient than GF104 while delivering much more raw performance and vastly more performance per watt. When SemiAccurate teases Nvidia's layout and physical design teams, it is for a reason.
Getting back to the story, to use ballpark figures, lets assume Cypress is now yielding at 75% for HD5870 and HD5850 combined, much higher if you add in HD5830. GF104 has one of it's seven shader groups fused off leaving 336 functional shaders, and upping yields. Lets just ballpark that chip's yield at 60% for reasons we can't get into. If Nvidia decides to release more GF104 derivatives, it will up the effective yield, but anything lower on the performance scale would go up against Juniper, a 166mm^2 chip. That is economic suicide.
This means you can get about 150 GF104 die candidates per 300mm wafer, at 60% yields, that is about 90 good parts. Cypress has about 160 candidates, and about 75% yields, meaning 120 good parts per wafer. You can play with the number as much as you like, see whatDavid Morgan did here for example, but the fact that Cypress is smaller and has notably higher yields is the take home message.
If a TSMC 40nm wafer costs $5000, that means each good 5850 or 5870 costs AMD about $42 to make. Each good GF104 costs Nvidia around $56. If you add in 5830s the number goes down for ATI, and the same holds true for any GF104 derivative that Nvidia releases. This is Pro-Ablem for reasons we will get into later, but not the end of the world for Nvidia.
The Cypress based HD5870 and HD5850 cards seem to be selling for about $380 and $300 at etail respectively while the GF104 based GTX460 sells for $199 and $229 with 768MB and 1GB of memory. If you assume that the card models are split 50-50 for supply, that means AMD cards sell for an average of $340 while Nvidia cards sell for $215 on average. The two Nvidia cards straddle the lowest end Cypress variant, the HD5830, in the market, and are well below the price and performance of the HD5850.
Getting back to the GPU prices, you might recall that Nvidia's silicon component costs about 25% more than AMD's, and the resultant card sells for about 63% of AMD's card sells for. Can you say squeezed margins, if they exist at all? ATI can lop $100 off their retail price and still make more money on Cypress than Nvidia currently makes on GF104.
Nvidia's will make some money on GF104 only as long as AMD feels like letting them. Because TSMC can't make enough 40nm wafers to go around, Cypress is currently supply constrained. Should GF104/GTX460 eat into AMD's marketshare enough to effectively allow Cypress demand to be met by current wafer availability, AMD will drop prices and likely push Nvidia into the red on GF104. It isn't a question of "Can AMD?", it is a question of "Will they?".
If you add in the low end HD5830 card, the numbers get better for AMD. ASPs go down, but so does cost because effective yield goes up. If Nvidia puts out a lower spec GF104 with 6 of the 8 clusters active, they will push yields up, but also push ASPs down a lot. That low end card would be competing with AMD's Juniper, a 166mm^2 part that costs far less than half of what Cypress costs to make. Once again, Nvidia is bringing a knife to 【请注意文明用词，试图绕过屏蔽将会被严厉处理】fight because their architecture is fundamentally broken.
If Nvidia puts out a higher spec GF104 with all 8 clusters active, they have the ability to raise ASPs, but yields will take a huge pounding, as will supply. This may be good for polishing the halo for the gullible onlookers, but it is doubtful it would help their financial situation much.
In the end, the GF104/GTX460 is a good card for consumers, it is priced right, and slots into a gaping hole in the AMD lineup. That hole exists because it is better for AMD's bottom line for it to be there. Nvidia poked a sleeping lion with the GTX460, and if it decides to roll over and swat the green team they have no defense.
To make matters worse, the Fermi architecture is not just wrong, it was also quarters late. About the time that Nvidia fleshes out the GF10x line with the upcoming GF106 and GF108 variants, AMD will have their next generation called Southern Islands out.
That will let AMD waterfall the prices on the current parts, and leave Nvidia without any place to turn, financially speaking. Best case, Nvidia has a window of a few months before AMD brings down the hammer. In the mean time, the boys in green have to be very careful, poking the sleeping lion might bring that hammer down faster. No matter which way they turn, Nvidia will get squished.S|A
作者: anolen01 时间: 2010-7-22 10:16
求翻译帝，msGF104核心面积367mm2？
作者: hdht 时间: 2010-7-22 10:16
提示: 作者被禁止或删除 内容自动屏蔽
作者: iamspy 时间: 2010-7-22 10:20
一篇国外的A枪的F.I.V.E. M.A.O.大作。
作者: clawhammer 时间: 2010-7-22 10:25
又是semiaccurate
作者: Sirlion 时间: 2010-7-22 10:31
说的nvidia因为这代费米要跨了似的
作者: WG_Baby 时间: 2010-7-22 10:34
召喚Edison和Asuka澄清吧
作者: 逆火 时间: 2010-7-22 10:53
重点是367mm²
作者: deadmeat 时间: 2010-7-22 11:02
提示: 作者被禁止或删除 内容自动屏蔽
作者: tangcw2002 时间: 2010-7-22 11:41
本帖最后由 tangcw2002 于 2010-7-22 11:42 编辑
die就算1000mm2又怎样，温度功耗在这里摆着的么。只要温度功耗合适，你做个10000mm2只要显卡pcb摆得下也无所谓，恩，10000mm2也不算大，不过10cm*10cm大小而已...
作者: alex_roy 时间: 2010-7-22 11:45
這篇文章是對的，我證明
作者: 风华绝代 时间: 2010-7-22 11:46
求翻译帝……看不明白，又抨击NV？
作者: AggressiveZone 时间: 2010-7-22 11:54
看那么大的王八盖子 估计小不了 不过价格还行 发热功耗都还好 你NV赔还是赚关玩家鸟事
作者: 380 时间: 2010-7-22 12:03
提示: 作者被禁止或删除 内容自动屏蔽
作者: eternal0 时间: 2010-7-22 12:06
NV赚与亏关玩家什么事，只要460性价比高就行了
作者: hdht 时间: 2010-7-22 12:06
提示: 作者被禁止或删除 内容自动屏蔽
作者: 66666 时间: 2010-7-22 12:18
满规格的GF100大家YY多久了，见到一**没有？？
做不出来就是做不出来，完整版再强，出不来的话他也是 ...
tft1122 发表于 2010-7-22 12:00 
你说这话不怕几个月后给人挖坟喷死？
完整版做不出来？
作者: qdamao 时间: 2010-7-22 12:26
想说什么啊，460不值得买还是啥的？
作者: geneme 时间: 2010-7-22 12:44
We can now tell you about GF104 die, it is 24.9mm x 14.7mm, and the die size is 367mm^2.
GF104的成本貌似不低
作者: higherren 时间: 2010-7-22 12:45
[quote]对了，辟谣的怎么还不进来呢？我都说了我说的估计是扯淡了
tft1122 发表于 2010-7-22 12:27 [/
精明，两头都说，基本将自己立于不败之地。但我猜你其实更希望这文章猜测的核心面积是事实吧？
作者: CoreSolo 时间: 2010-7-22 12:46
满规格104性能和RV870满规格也差不多，DX11略好。
darkstorm 发表于 2010-7-22 11:54
作者: Asuka 时间: 2010-7-22 12:48
semiaccurate.com的创始人是被inq扫地出门，才自己开的网站
在460发布当天，他连写了4篇文章黑NV
这种人的文字可信度要打一个很大的问号
作者: geneme 时间: 2010-7-22 12:54
全规格的GF104 第四季度出不来的话也就不用出了 貌似10月份amd 6系列第一批核心量产
作者: higherren 时间: 2010-7-22 12:58
回复 32# tft1122
其实对这我没什么希望的，只对分析某些有意思的内心世界感兴趣。
作者: 消失的古容ID 时间: 2010-7-22 13:01
10月份AMD 6系列的第一批核心量产？
PS：我以前的ID“古容”怎么禁止访问了？以前的帖子也都被删了。不像别人被禁言那样写到“此帖被版主或管理屏蔽”，而是直接消失了，我的那层楼也没了。。。貌似被禁言也不会删帖子吧。。
作者: nfsking2 时间: 2010-7-22 13:05
满规格的GF100大家YY多久了，见到一**没有？？
做不出来就是做不出来，完整版再强，出不来的话他也是 ...
tft1122 发表于 2010-7-22 12:00
厄，之前GF100在某些人口中也是完全做不出来，要推倒了重来，而他们口中的做不出来就是在我手中的GTX480，不知道是这些人穿越了还是我梦游了
作者: sosic 时间: 2010-7-22 13:16
200美刀不到的东西。。开个盖测一下不就行了。。搞这么多字累不累。。
作者: 古九寒 时间: 2010-7-22 13:34
那个查理的很多东西都他自己猜测的.
我只是想了解一下: 同一面积下晶体的密度比较高, yield 应该会降低吧?
作者: ifu 时间: 2010-7-22 13:38
面积无所谓，只要价格够便宜就行了
作者: 防转攻 时间: 2010-7-22 13:45
什么N卡用户A卡用户 便宜又好用的卡用户才会买 现在物理加速对绝大多数游戏影像不大 所以这个指标暂不考虑 性能价格差不多的情况下 多这么一个更好 少了也无所谓 哪个便宜买哪个 粉A的粉N的 **都是进水的 哪一家倒闭了 大家都没好果子吃 别人爱用什么卡用什么卡 你凭什么非得让人认同你的观点？
作者: los_parrot 时间: 2010-7-22 13:48
那个查理的很多东西都他自己猜测的.
我只是想了解一下: 同一面积下晶体的密度比较高, yield 应该会降低吧?
古九寒 发表于 2010-7-22 13:34
nV的shader要上2倍频率,高频部件的密度肯定要小一点.
==
367MMv2这个数字都出来了,肯定是有人开盖量了.而且那么大的盖,里面难道只有小小一块核心吗?这个数字上乱说的可能性不大.至少一上市,amd自己就会买过来开盖研究——假如他们在上市之前都还不知道的话。
SA的中心思想是gtx460对消费者来说是最佳的显卡,但是对nV来说情况不乐观.amd手握大锤。
下代制程上amd脚踩2条船，GF/tsmc两边都搞，万一tsmc不给力，nV的fermi悲剧就会重演了。
作者: kakaku.bj.cn 时间: 2010-7-22 14:13
求翻译帝，msGF104核心面积367mm2？
anolen01 发表于 2010-7-22 10:16
1. 之前爆的核心366mm^2并不是错误，实际是367,techreport上的320才是错的，而nv并不想让人知道这一点。
2. 比对手58XX大10%，但晶体管数少0.2个十亿，amd塞晶体管的技术强20%
3. 通过一堆计算说每个好用的5850/70成本$42, 每个好用的gf104成本$56,但是最终卖价远低于cypress,
4. gf106/108的对手将是SI
以上纯lz文大意，真伪不知
作者: 排骨饭 时间: 2010-7-22 14:42
看来460还真的是对一部分人造成了很大的精神压力。
作者: chery66 时间: 2010-7-22 14:45
为什么我觉得semi到现在为止的爆料就没有失败过...
作者: 什么？ 时间: 2010-7-22 15:02
semiaccurate.com的创始人是被inq扫地出门，才自己开的网站
在460发布当天，他连写了4篇文章**V
这种 ...
Asuka 发表于 2010-7-22 12:48
此人虽然**V，但历史上看，预测的准确率都蛮高的，比通常意义上的预言家高很多。
这次有板有眼，黑纸白字的写了367mm2，相比是靠谱的
作者: disruptor 时间: 2010-7-22 15:11
这个面积确实挺大的。。。怪不得460的公版那么差的做工
作者: kevinkt 时间: 2010-7-22 15:27
有個傻子看著別人**,自己就跟著高潮了
作者: LockheedMartin 时间: 2010-7-22 15:46
提示: 作者被禁止或删除 内容自动屏蔽
作者: moshou 时间: 2010-7-22 15:48
本帖最后由 moshou 于 2010-7-22 15:50 编辑
看来460还真的是对一部分人造成了很大的精神压力。
排骨饭 发表于 2010-7-22 14:42
+1
咱用卡的人可不管GF104的芯片有多长多宽多重多厚，只要价钱便宜性能够爽功耗适度温度清凉它就是块好卡
就算大10%又怎么的
作为同档价格的游戏卡，DX1......DX11一样不缺性能一样不差，DX11游戏性能还更强，DX11 MS SDK DEMO成几倍几十倍的屠杀HD58XX（某A语录：抱ATI大腿的MS的DX11标准才是游戏的未来）
还可享受PHSYX游戏、CUDA后处理的游戏，本人觉得大10%还少了，应该大更多
至于HD58XX，在本人看来，作为其标榜的DX11游戏卡，相近价格，却提供稀烂的DX11性能、部分游戏中无法提供PHSYX/CUDA后处理，这样一块游戏功能性能残废卡，少了那么多东西，只小10%？太浪费晶体管了
个人结论：HD58XX就是NV30加强版！
相对当年的NV30,不同点是：
58XX比同代对手有先发布半年优势，功耗温度比对手GF100好，NV30则相反。
游戏性能方面，58XX由于先发布老版DX不如同代对手，NV30则老DX略胜同代对手OPENGL大胜同代对手
游戏功能方面，58XX部分游戏PHSYX、CUDA后处理特效不支持，该卡用户只能忍痛选择无视此类游戏，因为他们无法接受自己的卡在此类游戏中被价格只有其几分之一的卡灭掉！而NV30相对同代对手老游戏功能无缺失
相同点是：
58XX与NV30所标称的新DX都是残废（按MS的标准，58XX更残废！MS SDK中只有对手几分之一甚至几十分之一的性能）
至于专业方面，相对同代对手，58XX给NV30提鞋都不配
作者: 古九寒 时间: 2010-7-22 16:50
看到标题我笑了.
367mm^2 vs 334mm^2
10%的面积差距, 这也叫Huge?
作者: moshou 时间: 2010-7-22 18:21
200美刀不到的东西。。开个盖测一下不就行了。。搞这么多字累不累。。
sosic 发表于 2010-7-22 13:16
对头，不就开个盖嘛
再说104die大小干我们普通用户啥事？怎么有些人急得上了火似的？
作者: 3118595 时间: 2010-7-22 19:56
如果说绝对做不出来是夸张了点，只不过是量的问题罢了，做多做少都是做嘛，赚个吆喝也不错。
但是等你 ...
tft1122 发表于 2010-7-22 12:24
7 ?! ^! R1 j( G# y8 `5 }% N. g, @ S7 y6 z
如果说绝对做不出来是夸张了点，只不过是量的问题罢了，做多做少都是做嘛，赚个吆喝也不错。
, [& v9 D# H ?+ M% @$ t$ j但是等你积了半年挤出来那点完整版核心又有什么意义呢？？( y0 N/ V% D( m. ?
市场早就被人家吃的干干净净了，而且说不定还要面对人家下一代的芯片.....1 q* V" J- f8 ?/ M ^5 H$ K
AMD吃了这么久 也没吃过50% 啥回事？AMD胃口太小？慈善 不想吃太多？
作者: 破灭时空 时间: 2012-2-7 12:12
这个帖子也需要挖一下，查理的GK104核心面积贴来了，时隔一年半之久
作者: recruitbj 时间: 2012-2-8 09:02
一大篇英文，呵呵
把主要意思翻一翻再发出来呗
作者: SnakeLee 时间: 2012-2-8 12:59
呵呵厚，这个有意思，不过是坟
| 欢迎光临 POPPUR爱换 (https://we.poppur.com/) |
Powered by Discuz! X3.4 |