anandtech ZEN PPT 分析文章中文翻譯 - 3C

Audriana avatar
By Audriana
at 2016-08-22T13:06

Table of Contents

看到對岸有人翻譯就轉過來了

有做部分名詞更動像是 緩存=>快取 多線程=>多執行緒

--
【AMD Zen 微架構完全解析:雙調度器,微指令快取以及快取架構】
AMD Zen Microarchitecture: Dual Schedulers, Micro-Op Cache and Memory
Hierarchy Revealed

原文:http://www.anandtech.com/show/10578/amd-zen-microarchitecture-dual-schedulers-micro-op-cache-memory-hierarchy-revealed

縮網址:http://goo.gl/BMfq8u

翻譯出處:http://www.mykancolle.com/?post=385

In their own side event this week, AMD invited select members of the press
and analysts to come and discuss the next layer of Zen details. In this
piece, we’re discussing the microarchitecture announcements that were made,
as well as a look to see how this compares to previous generations of AMD
core designs.

AMD這邊邀請了部分媒體和分析師參與Zen的進一步細節的討論。 這篇文章裡我們將討
論架構,並與前代處理器作比較。

--
Prediction, Decode, Queues and Execution【分支預測、解碼、佇列以及執行】

First up, let’s dive right into the block diagram as shown:

首先讓我們之間看下面的圖

http://images.anandtech.com/doci/10578/s1%20Perf.png

If we focus purely on the left to start, we can see most of the high-level
microarchitecture details including basic caches, the new inclusion of an
op-cache, some details about decoders and dispatch, scheduler arrangements,
execution ports and load/store arrangements. A number of slides later in the
presentation talk about cache bandwidth.

如果我們從左邊開始看起,我們可以看到大部分的架構細節,包括L1快取、新的微指令
快取、解碼、分發、調度器、執行埠以及L/S單元的設計。

Firstly, one of the bigger deviations from previous AMD microarchitecture
designs is the presence of a micro-op cache (it might be worth noting that
these slides sometimes say op when it means micro-op, creating a little
confusion). AMD’s Bulldozer design did not have an operation cache,
requiring it to fetch details from other caches to implement frequently used
micro-ops. Intel has been implementing a similar arrangement for several
generations to great effect (some put it as a major stepping stone for
Conroe), so to see one here is quite promising for AMD. We weren’t told the
scale or extent of this buffer, and AMD will perhaps give that information in
due course.

首先,Zen與前代架構的很大一處不同在於,出現了微指令快取(ppt上有時候寫的op快
取,實際上意思就是micro-op,容易誤導人)。 AMD的推土機設計沒有微指令快取,就必
須從其他快取中提取細節,來執行頻繁使用的微指令。 Intel很早就開始用微指令快取了
,效果非常好(在Conroe架構上引入的重要改進),所以對於AMD來說這應該能帶來不小
提升。 AMD沒告訴我們這個緩衝區的大小,估計在適當的時候會給出資訊。

Aside from the as-expected ‘branch predictor enhancements’, which are as
vague as they sound, AMD has not disclosed the decoder arrangements in Zen at
this time, but has listed that they can decode four instructions per cycle to
feed into the operations queue. This queue, with the help of the op-cache,
can deliver 6 ops/cycle to the schedulers. The reasons behind the queue being
able to dispatch more per cycle is if the decoder can supply an instruction
which then falls into two micro-ops (which makes the instruction vs micro-op
definitions even muddier). Nevertheless, this micro-op queue helps feed the
separate integer and floating point segments of the CPU. Unlike Intel who
uses a combined scheduler for INT/FP, AMD’s diagram suggests that they will
remain separate with their own schedulers at this time.

拋開含糊不清的「增強的分支預測器」,AMD這次也沒披露解碼器的設計,但列出他們
每週期可以解碼4條指令到佇列。 這個佇列在微指令快取的輔助下,到調度器時能達到最
高每週期6條指令。 因為解碼器可以解碼一條指令,然後該指令隨後拆分為兩條微指令(
這讓指令和微指令的區別變得模糊)。 此外,這個微指令佇列還能提高每個整數和浮點
單元的利用率。 AMD不像Intel那樣給整數/浮點一個公用的調度器,而是繼續使用分離的
調度器。

The INT side of the core will funnel the ALU operations as well as the
AGU/load and store ops. The load/store units can perform 2 16-Byte loads and
one 16-Byte store per cycle, making use of the 32 KB 8-way set associative
write-back L1 Data cache. AMD has explicitly made this a write back cache
rather than the write through cache we saw in Bulldozer that was a source of
a lot of idle time in particular code paths. AMD is also stating that the
load/stores will have lower latency within the caches, but has not explained
to what extent they have improved.

整數部分包括ALU、AGU以及LS操作。 LS單元每週期可以執行2次16位元組的load以及1
次16位元組的store操作,利用32KB 8路組相連 回寫式L1資料快取。 AMD明確說明這是回
寫式快取,而不是推土機上的穿透式快取(在一定條件下會帶來大量的閒置時間)。 AMD
聲稱快取內的LS操作延遲會更低,但沒再做進一步說明。

The FP side of the core will afford two multiply ports and two ADD ports,
which should allow for two joined FMAC operations or one 256-bit AVX per
cycle. The combination of the INT and FP segments means that AMD is going for
a wide core and looking to exploit a significant amount of instruction level
parallelism. How much it will be able to depends on the caches and the
reorder buffers – no real data on the buffers has been given at this time,
except that the cores will have a +75% bigger instruction scheduler window
for ordering operations and a +50% wider issue width for potential
throughput. The wider cores, all other things being sufficient, will also
allow AMD’s implementation of simultaneous multithreading to potentially
take advantage of multiple threads with a linear and naturally low IPC.

每核心浮點部分包括兩個乘法埠,兩個ADD埠,每週期能夠執行兩條捆綁的FMAC命令或
者一條256bit AVX。 把整數和浮點部分合起來看,Zen核心在指令級並行上將會有很大提
升 - 提升多少取決於快取和重排序快取 - 這次沒給出ROB的具體資料,只說排序操作的
指令調度視窗將會增大75%,發射寬度提升50%。 即便是天生IPC就低的AMD處理器,核心
並行性越好,其他的方面就有效率多了,這也使得這次用的SMT在多執行緒上占得先機。


Deciphering the New Cache Hierarchy 【解密新的快取結構】

http://images.anandtech.com/doci/10578/s3%20Cache.png

The cache hierarchy is a significant deviation from recent previous AMD
designs, and most likely to its advantage. The L1 data cache is both double
in size and increased in associativity compared to Bulldozer, as well as
being write-back rather than write-through. It also uses an asymmetric
load/store implementation, identifying that loads happen more often than
stores in the critical paths of most work flows. The instruction cache is no
longer shared between two cores as well as doubling in associativity, which
should decrease the proportion of cache misses. AMD states that both the L1-D
and L1-I are low latency, with details to come.

這次的快取結構相比以前做出了重大改進,而且是朝著好的方向。 相較于推土機,Zen
的L1快取在大小和關聯性都翻倍了,而且是寫回式而不是穿透式。 同時採用了非對稱LS
單元,因為在大多數情況下Load操作比Store要頻繁得多。 指令快取不再是兩個核心共用
,同時關聯性也翻倍,這將減少快取未命中的情況。 AMD聲稱L1資料和指令快取延遲都很
低,今後將公佈更多細節。

The L2 cache sits at half a megabyte per core with 8-way associativity, which
is double that of Intel’s Skylake which has 256 KB/core and is only 4-way.
On the other hand, Intel’s L3/LLC on their high-end Skylake SKUs is at 2
MB/core or 8 MB/CPU, whereas Zen will feature 1 MB/core and both are at
16-way associativity.

L2快取變成了每核心512KB,8路相連,這是Intel Skylake上256kb 4路關聯的兩倍。
另一方面,Intel的L3在高端Skylake i7上是每核心2MB,每CPU8MB,在Zen上則是每核心
1MB,這兩者都是16路關聯。

Edit 7:18am: Actually, the slide above is being slightly evasive in its
description. It doesn't say how many cores the L3 cache is stretched over, or
if there is a common LLC between all cores in the chip. However, we have
recieved information from a source (which can't be confirmed via public AMD
documents) that states that Zen will feature two sets of 8MB L3 cache between
two groups of four cores each , giving 16 MB of L3 total. This would means 2
MB/core, but it also implies that there is no last-level unified cache in
silicon across all cores, which Intel has. The reasons behind something like
this is typically to do with modularity, and being able to scale a core
design from low core counts to high core counts. But it would still leave a
Zen core with the same L3 cache per core as Intel.

實際上上面的ppt在描述上有點曖昧。 沒有說多少核心共用8M L3,更沒說是否每顆晶
片上的所有核心都是共用同一個L3的。 然而我們從一個消息來源獲得的資訊(在AMD官方
ppt上找不到的)表明,Zen的8核晶片上是4個核心為一個簇,每個簇4個核心共用8M L3,
8核晶片有兩組8MB,共16MB的 L3。 這樣的話就是每核心2MB,但這也說明了Zen的L3不是
完全共用的,然而Intel的是完全共用的。 這樣做的原因估計和模組化有點關係,通過增
加這樣的模組可以做出從4核心直到32核心,但Zen的每核心L3和Intel的依然都是每核心
2MB(沒有任何優勢)

http://i.imgur.com/JSBgrgJ.jpg

What this means, between the L2 and the L3, is that AMD is putting more lower
level cache nearer the core than Intel, and as it is low level it becomes
separate to each core which can potentially improve single thread
performance. The downside of bigger and lower (but separate) caches is how
each of the cores will perform snoop in each other’s large caches to ensure
clean data is being passed around and that old data in L3 is not out-of-date.
AMD’s big headline number overall is that Zen will offer up to 5x cache
bandwidth to a core over previous designs.

這也意味著,AMD的L1和L2比Intel更大、延遲更低。 而且L1、L2距離核心更近,還是
每核心獨立的,在單線程性能上會有顯著提升。 但更大的獨立L1/L2帶來的壞處是,每個
核心都要監聽其他核心的快取,確保 1.傳遞的是乾淨資料、2. L3上的原資料不過期。
AMD給出的總體數位是,Zen在快取頻寬上是前代的5倍。


Low Power, FinFET and Clock Gating 【低功耗,FinFET,門控時鐘】

When AMD launched Carrizo and Bristol Ridge for notebooks, one of the big
stories was how AMD had implemented a number of techniques to improve power
consumption and subsequently increase efficiency. A number of those lessons
have come through with Zen, as well as a few new aspects in play due to the
lithography.

在AMD發佈Carrizo和Bristol Ridge的時候,介紹的一個重點就是一系列降低功耗和提
升能效的技術。 有一部分技術延續到了Zen上,同時伴隨著制程更新,還加入了一些新的
技術。

http://images.anandtech.com/doci/10578/s5%20FinFET.png

First up is the FinFET effect. Regular readers of AnandTech and those that
follow the industry will already be bored to death with FinFET, but the
design allows for a lower power version of a transistor at a given frequency.
Now of course everyone using FinFET can have a different implementation which
gives specific power/performance characteristics, but Zen on the 14nm FinFET
process at Global Foundries is already a known quantity with AMD’s Polaris
GPUs which are built similarly. The combination of FinFET with the fact that
AMD confirmed that they will be using the density-optimised version of 14nm
FinFET (which will allow for smaller die sizes and more reasonable efficiency
points) also contributes to a shift of either higher performance at the same
power or the same performance at lower power.

首先就是FinFET。 雖然大部分的人都已經熟悉FinFET到吐了,但我們還是要介紹一下
。 FinFET設計能在給定頻率下設計出電晶體的低功耗版本。 每個FinFET代工廠給出的技
術指標都不同,但Zen用的GF 14nm技術和Polaris GPU的不會差太多,這意味著AMD使用的
是14nm的追求密度版本,能在同等功耗下達成更高性能,或者低功耗下達成同等性能。

http://images.anandtech.com/doci/10578/s6%20Efficiency.png

AMD stated in the brief that power consumption and efficiency was constantly
drilled into the engineers, and as explained in previous briefings, there
ends up being a tradeoff between performance and efficiency about what can be
done for a number of elements of the core (e.g. 1% performance might cost 2%
efficiency). For Zen, the micro-op cache will save power by not having to go
further out to get instruction data, improved prefetch and a couple of other
features such as move elimination will also reduce the work, but AMD also
states that cores will be aggressively clock gated to improve efficiency.

AMD介紹說工程師們一直很注重功耗和能效,在性能和功能單元的效率上做了很多權衡
(比如提升1%的性能,代價是2%的能效損失)。 不僅有微指令快取可以節約讀取指令快
取的電能,改善的預取機制等也能減少工作量。 但AMD也說明,為了提升能效,Zen的門
控時鐘將會很激進。

We saw with AMD’s 7th Gen APUs that power gating was also a target with that
design, especially when remaining at the best efficiency point (given
specific performance) is usually the best policy. The way the diagram above
is laid out would seem to suggest that different parts of the core could
independently be clock gated depending on use (e.g. decode vs FP ports),
although we were not able to confirm if this is the case. It also relies on
having very quick (1-2 cycle) clock gating implementations, and note that
clock gating is different to power-gating, which is harder to implement.

AMD第七代APU上也有差不多的設計,保持在效率最高的那個點(特定性能)是最好的方
式。 上圖似乎暗示著每個核心的不同部分(取決於用途)都有獨立的門控時鐘(比如解
碼單元或者浮點埠),雖然目前還無法確認。 同時還需要有非常快速的門控時鐘(1-2個
週期),要知道門控時鐘與功耗門限不同,門控時鐘更難設計。


Simultaneous Multi-Threading 【同步多執行緒】

On Zen, each core will be able to support two threads in what is called ‘
simulatenous multi-threading’. Intel has supported their version of SMT for
a number of years, and other CPU manufacturers like IBM support up to 8
threads per core on their POWER8 platform designs. Building a core to be able
to use multiple threads can be tough, as it requires a lot of resources to
make sure that the threads do not block each other by consuming all the cache
and buffers in play. But AMD will equip Zen with SMT which means we will see
8C/16T parts hitting the market.

Zen架構上,每個核心支援兩個執行緒,這叫做同步多執行緒。 Intel版本的SMT早在08
年就開始啟用,其他的廠商比如IBM,在POWER8上支援最多8個執行緒(SMT8)。 讓一個
核心處理兩個執行緒很困難,需要很多資源來確保執行緒之間不會因爭奪快取而互相阻塞
。 Zen桌上出版將會有8核16執行緒。
http://images.anandtech.com/doci/10578/s4%20SMT.png

Unlike Bulldozer, where having a shared FP unit between two threads was an
issue for floating point performance, Zen’s design is more akin to Intel’s
in that each thread will appear as an independent core and there is not that
resource limitation that BD had. With sufficient resources, SMT will allow
the core instructions per clock to improve, however it will be interesting to
see what workloads will benefit and which ones will not.
在推土機上,共用浮點單元使得浮點性能不如人意。 但Zen的設計更類似于Intel,每
個執行緒都和一個單獨核心差不多,不會有推土機上的資源限制。 有了更多的資源,SMT
將會提升IPC,我很想看看哪些負載能從中獲益。 】


Timeframe and Availability 【時程表、供貨日期】

At the presentation, it was given that Zen will be available in volume in
2017. As the AM4 platform will share a socket with Bristol Ridge, users are
likely to see Bristol Ridge systems from AMD’s main OEM partners, like Dell
and others, enter the market before separate Zen CPUs will hit the market for
DIY builders. It’s a matter of principle that almost no consumer focused
semiconductor company releases a product for the sale season, and Q1 features
such events as CES, which gives a pretty clear indication of when we can
expect to get our hands on one.

在ppt上寫著Zen將會在17年大量出貨。 由於AM4平臺上Summit Ridge和Bristol Ridge
使用同樣的插槽,可能我們能從AMD的OEM們那裡先見到Bristol Ridge進入市場。 沒有哪
個主賣消費級產品的半導體廠商會在年末清倉季發佈新品,而第一季度會有CES之類的大
型展會,那時候我們肯定能拿到手。

It’s worth noting that AMD said that as we get closer to launch, further
details will come as well as deeper information about the design. It was also
mentioned that the marketing strategy is also currently being determined,
such that Zen may not actually be the retail product name for the line of
processors (we already have Summit Ridge as the platform codename, but that
could change for retail as well).

AMD說距離發佈越近,就會公佈越多的架構細節。 還提到了行銷策略上的決定,比如
Zen不會是實際產品線的名稱(實際平臺代號是Summit Ridge,但到了出貨時候也可能會
變)。


Wrap Up 總結

AMD has gone much further into their core design than I expected this week.
When we were told we had a briefing, and there were 200-odd press and
analysts in the room, I was expecting to hear some high level puff about the
brand and a reiteration of their commitment to the high end. To actually get
some slides detailing parts of the microarchitecture, even at a basic cache
level, was quite surprising and it somewhat means that AMD might have stolen
the show with the news this week.

這次AMD的介紹比我想的要深入。 當有人告訴我去參加一個短會,並且會有200多家媒
體和分析師到場,我還估計應該就是吹吹牛逼,重申要回到高端市場什麼的。 但實際上
AMD給出了部分架構的詳盡介紹,甚至還介紹了基本快取結構,這出乎我的意料,估計這
個星期媒體上都會是AMD的新聞了吧。


--
Tags: 3C

All Comments

Jacob avatar
By Jacob
at 2016-08-24T03:59
有換名詞給推
Donna avatar
By Donna
at 2016-08-24T08:26
為什麼好好的要吹牛逼
Caitlin avatar
By Caitlin
at 2016-08-25T19:09
這周不是有Hotchips的活動嗎 會有更多ZEN的消息吧
Suhail Hany avatar
By Suhail Hany
at 2016-08-27T18:25
樓上要加碼股票嗎XD
Lily avatar
By Lily
at 2016-08-29T10:13
PTT又來拉XD
PPT
Susan avatar
By Susan
at 2016-09-02T01:10
大陸網友評論:「可以靠臉蛋,卻偏偏要靠才華。」
Christine avatar
By Christine
at 2016-09-04T12:21
Intel從Meron開始,一週期4個decoder執行4uop,一
直到skylake才改成一週期5decoder執行6uop,硬是拉
高一些效能,先看看AMD怎麼搞…
Candice avatar
By Candice
at 2016-09-06T01:19
等到2017年還在PCI-E 2.0 NVMe直接無視就對了
Agatha avatar
By Agatha
at 2016-09-10T16:27
板主在爆AM4的掛?
Mia avatar
By Mia
at 2016-09-11T15:35
有掛!
Hamiltion avatar
By Hamiltion
at 2016-09-14T07:23
這很多人都知道啊 那顆fch....QQ
Elma avatar
By Elma
at 2016-09-15T08:22
版主 怎麼會知道 am4的掛?
Iris avatar
By Iris
at 2016-09-18T15:53
因為AM4的晶片組據說是外包給祥碩
電蝦板很多業內人在看啊
Genevieve avatar
By Genevieve
at 2016-09-18T23:22
版主又不是普通人XD
Puput avatar
By Puput
at 2016-09-19T13:00
板主每個也來頭不小XD
Zora avatar
By Zora
at 2016-09-23T18:34
喔!
Ina avatar
By Ina
at 2016-09-28T04:56
該不會CPU看到車尾燈,結果死在晶片組吧QQ
Genevieve avatar
By Genevieve
at 2016-09-29T12:20
又一個被豬隊友害死的例子
Hedda avatar
By Hedda
at 2016-10-02T16:42
amd 加油 intel 都在裝死
Kristin avatar
By Kristin
at 2016-10-05T05:58
還好我買不起NVMe SSD 無感
Olga avatar
By Olga
at 2016-10-06T20:55
聽說某外包公司賠惹不少錢QQ
Susan avatar
By Susan
at 2016-10-07T16:27
結果晶片組烙賽
Mary avatar
By Mary
at 2016-10-11T16:49
那間A公司好像i社也賠 a社也賠@@
amd之後有錢惹大概就收回來自己做惹
Edith avatar
By Edith
at 2016-10-11T20:36
http://i.imgur.com/483hJvR.jpg 雷包晶片?
Faithe avatar
By Faithe
at 2016-10-13T12:11
各種賠
Regina avatar
By Regina
at 2016-10-13T16:08
FM2+的不是有PCI-E 3.0? 不能拿FM2+的改AM4腳位嗎
Andy avatar
By Andy
at 2016-10-15T23:57
自己做AM4+嗎
David avatar
By David
at 2016-10-20T19:58
fm2+的pcie3應該是cpu上的八
那葛zen也有喇
Yuri avatar
By Yuri
at 2016-10-23T08:43
好久不見的先知!
Catherine avatar
By Catherine
at 2016-10-26T16:46
AMD 的Gen3 不就是obov做的嗎?
Annie avatar
By Annie
at 2016-10-30T09:16
zen跟ploaris上的不是我做的惹QQ
Kyle avatar
By Kyle
at 2016-11-01T05:42
雖然說是拿我的20nm去port到14QQ
Annie avatar
By Annie
at 2016-11-05T14:45
沒跟到zen這顆革命性cpu真的有點可惜QQ
不過水果行一年200M隻手機感覺也是葛秋QQ
Skylar DavisLinda avatar
By Skylar DavisLinda
at 2016-11-06T11:52
教主的20nm (蓋章
Regina avatar
By Regina
at 2016-11-07T22:19
我的30cm(誤
Skylar DavisLinda avatar
By Skylar DavisLinda
at 2016-11-10T16:38
http://i.imgur.com/v2kgv28.jpg 開這種鳥規格
Delia avatar
By Delia
at 2016-11-13T07:19
教主趕快回AMD拯救世界
Ida avatar
By Ida
at 2016-11-14T21:40
拖到 2017 沒有 pcie 3.0 的 io hub 真可憐
Agnes avatar
By Agnes
at 2016-11-19T03:04
一開始看到那張投影片我還以為我眼花了
Rebecca avatar
By Rebecca
at 2016-11-20T19:56
哀.........
水果行有他機掰的地方 也有她迷人的地方
Lucy avatar
By Lucy
at 2016-11-21T02:27
靠! 真的 只有 PCIe 2.0喔
Eden avatar
By Eden
at 2016-11-22T17:43
教主只有20nm (蓋章
Jacky avatar
By Jacky
at 2016-11-24T20:51
全世界有幾億人再用你做的東西OOXX 那種感覺
在amd很難QQ
顯卡pcie直接cpu的
Una avatar
By Una
at 2016-11-26T02:57
教主的20nm被做成14nm
Linda avatar
By Linda
at 2016-11-29T22:16
顯卡的PCIE是通南橋還是CPU啊 @@?
Connor avatar
By Connor
at 2016-12-01T08:52
PCIE的SSD先普及再說 目前有幾台電腦配那種SSD
Frederica avatar
By Frederica
at 2016-12-06T00:46
顯卡一般還是通cpu啦
Margaret avatar
By Margaret
at 2016-12-06T08:52
南橋說實在沒很重要喇
Mary avatar
By Mary
at 2016-12-08T06:50
顯卡PCIE是走CPU的 儲存裝置才是走FCH
Jacob avatar
By Jacob
at 2016-12-08T21:50
PCIe影響的又不是只有SSD
Frederica avatar
By Frederica
at 2016-12-09T20:53
南橋一般的pcie都拿去接其他東西
storage也可以直接上北橋/cpu
Dinah avatar
By Dinah
at 2016-12-10T23:25
m8pe版上一堆開箱 會沒很多人買嗎
而且現在是2017年了耶
打錯是要2017年推出耶
Elvira avatar
By Elvira
at 2016-12-11T19:22
先知一出就是猛料!
Frederica avatar
By Frederica
at 2016-12-13T21:12
Zen上面給顯卡吃的還是PCIE3.0阿 那是做在CPU裡的
Ida avatar
By Ida
at 2016-12-15T20:56
感覺教主在AMD各種心酸
Selena avatar
By Selena
at 2016-12-17T01:23
其實一般消費級南橋只要SATAIII跟USB3.0開的夠
Emily avatar
By Emily
at 2016-12-21T13:22
就沒太多人會嫌了
要gen3 NMVe的可能也算高端玩家了
Edwina avatar
By Edwina
at 2016-12-25T03:07
2017應該是要有usb3.1 QQ
我覺得某A公司真的 哀算惹
Mason avatar
By Mason
at 2016-12-25T10:15
8條尬 24條 接幾摳咖稱
James avatar
By James
at 2016-12-29T21:35
伺服器市場不是還蠻注重IO性能的嗎
Lydia avatar
By Lydia
at 2016-12-30T22:23
買到便宜的M8PeG 路過
Olga avatar
By Olga
at 2017-01-04T00:21
我也這麼覺得 開遊戲又沒差載入那幾秒
Ivy avatar
By Ivy
at 2017-01-07T05:52
買到促銷的M8PeG 路過
Noah avatar
By Noah
at 2017-01-09T01:30
整個qq 都醞釀這麼久了結果fch還是只有那樣
也不用說高端啦 現在一堆台廠控制器下來玩
John avatar
By John
at 2017-01-09T07:52
台廠技術能力還是差美國很大一截
gen2 ip好做又便宜阿 gen3價錢三倍五倍都有
Thomas avatar
By Thomas
at 2017-01-12T12:14
坐等AM4+ (?
Tracy avatar
By Tracy
at 2017-01-13T03:37
U3.1還可以外掛啦 系統廠cost up就是了
Kristin avatar
By Kristin
at 2017-01-14T23:52
看教主買AMD的股票,不買AMD的產品 大概就了解了QQ
Caroline avatar
By Caroline
at 2017-01-19T07:25
上週沒買到M8PeG 漲了 QQ
Freda avatar
By Freda
at 2017-01-21T14:45
所以 am4 就算沒支援3.0 cpu有支援3.0就夠摟
Lily avatar
By Lily
at 2017-01-23T01:26
沒錯啊
Lucy avatar
By Lucy
at 2017-01-26T18:36
還有主機板上的PCI-E頻寬Switch會不會突然衝康
Doris avatar
By Doris
at 2017-01-28T20:03
U3S6的頻寬對於大部分的用戶根本還嫌多
William avatar
By William
at 2017-01-28T22:20
哦!
Ivy avatar
By Ivy
at 2017-01-31T18:09
我去水果行前 還有人在找我談兼差做usb3.1
Skylar Davis avatar
By Skylar Davis
at 2017-02-05T07:15
後來算一算沒很好賺 算惹QQ
Sierra Rose avatar
By Sierra Rose
at 2017-02-06T09:24
只是這此不是說有重返高端 市場 假如沒支援 nvme 介
面 會不會 功虧一簣啊?
Bennie avatar
By Bennie
at 2017-02-07T10:42
QQ
反正X79沒支援NVMe 買了也是浪費 組SATA RAID0就好
Damian avatar
By Damian
at 2017-02-08T17:48
這根本不是夠不夠的問題啊w 你旗艦產品作一個
trash spec..這有點說不過去吧
Faithe avatar
By Faithe
at 2017-02-09T06:46
認真的教主
Emma avatar
By Emma
at 2017-02-12T12:35
我在amd的同事是跟我說再做pcie gen4惹
Cara avatar
By Cara
at 2017-02-13T22:24
amd又沒有急著賣旗艦的價錢 股東角度沒差喇
會南僑外包給台灣的A公司也是權宜之計QQ
Ula avatar
By Ula
at 2017-02-17T11:49
gen gen der
Belly avatar
By Belly
at 2017-02-22T06:26
所以又要再等一代?等amd收回去自己做?
Oscar avatar
By Oscar
at 2017-02-25T20:52
對啦 但對一般升斗小民來說都夠用也是事實
偏偏這些升斗小民的電腦才是市場上的主流
Kristin avatar
By Kristin
at 2017-02-27T02:16
南橋那種東西大概不用一年就搞定惹
需要的ip amd自己都有 就soc接一接tapeout
Rachel avatar
By Rachel
at 2017-02-27T07:01
http://i.imgur.com/zWymNUb.jpg 直接衝康 UMI link
Zanna avatar
By Zanna
at 2017-03-03T18:33
看來AMD真的除ZEN以外其他的都砍光光了
Bennie avatar
By Bennie
at 2017-03-07T03:00
測三葛月送es再三葛月上市
Tom avatar
By Tom
at 2017-03-10T06:28
Intel也是到skylake才放gen3在pch 這很嚴重嗎?
Rachel avatar
By Rachel
at 2017-03-11T23:58
所以2017末會出am4+之類的?
Kumar avatar
By Kumar
at 2017-03-14T14:48
AM4不是10月就要出貨了,還來得及改嗎?
Daph Bay avatar
By Daph Bay
at 2017-03-18T00:07
有時候只是缺個理由XD
Ethan avatar
By Ethan
at 2017-03-22T22:22
死在晶片組 QQ
Necoo avatar
By Necoo
at 2017-03-26T07:03
QQ
Thomas avatar
By Thomas
at 2017-03-26T16:22
是那個 a開頭 的愛代理商 在作的晶片?
Annie avatar
By Annie
at 2017-03-30T14:14
metal change一葛月到一葛半
Quanna avatar
By Quanna
at 2017-04-04T00:31
不過ZEN真的不能再延了 只有PCIe 2.0也要硬著頭皮上
Necoo avatar
By Necoo
at 2017-04-04T12:54
Zen有PCIE3.0阿 講的好像沒有一樣
Connor avatar
By Connor
at 2017-04-04T22:06
越想越覺得台廠豬屎屋真的有點糟QQ
Audriana avatar
By Audriana
at 2017-04-09T01:25
那個某a很...豬隊友ㄚ
Agnes avatar
By Agnes
at 2017-04-11T22:53
我很擔心amd收回去做晶片組會不會順便改腳位阿 這樣
第一代不就跟passcard一樣下場
Delia avatar
By Delia
at 2017-04-12T17:23
就只是缺個理由啊 某人愛黑AMD又不是新聞XD
Kyle avatar
By Kyle
at 2017-04-14T16:01
講錯 a開頭的代理商 作的?
Madame avatar
By Madame
at 2017-04-15T00:49
那個全世界也只有Intel有做起來而已不是QQ
Skylar DavisLinda avatar
By Skylar DavisLinda
at 2017-04-15T05:12
不可能喇 amd哪那摸白痴
Delia avatar
By Delia
at 2017-04-19T12:04
不過祥碩出那麼大的包怎麼沒反應在股價上?
Heather avatar
By Heather
at 2017-04-24T07:51
一開始不是就說fch不行了 哪來說zen沒pcie3
David avatar
By David
at 2017-04-29T02:13
沒人去宣傳阿
Charlie avatar
By Charlie
at 2017-04-29T08:23
760之後好久沒買過AMD晶片組了XD
Gilbert avatar
By Gilbert
at 2017-05-02T09:23
amd自己來大概還不會這麼悲劇
John avatar
By John
at 2017-05-07T03:59
應該說就算AM4只有PCIe2.0 為了Zen也要硬上啦
Gilbert avatar
By Gilbert
at 2017-05-07T18:17
就南僑整組裁惹 director改掛fellow
現在在加拿大狂招人
Una avatar
By Una
at 2017-05-09T11:12
ZEN跟最初的預期延了快一年了吧
Ivy avatar
By Ivy
at 2017-05-13T10:08
不會啦 顯卡走CPU直接出來的
少量出貨 這招AMD又不是第一次用 還有人上鉤QQ
Connor avatar
By Connor
at 2017-05-15T03:23
Zen當初不就是瞄準年底聖誕節嗎?
翻一下教主之前的文就有講了
Odelette avatar
By Odelette
at 2017-05-20T02:36
那葛規格應該外包給某A的時候就談好惹
Puput avatar
By Puput
at 2017-05-20T13:16
gen3ip貴
而且我也很懷疑台廠有多少gen3的經驗跟能力
ati的大本營阿
以前喵喵大部分都ati的人做的
Una avatar
By Una
at 2017-05-24T15:46
搬到加拿大到底想做啥啊 我實在不懂AMD
Todd Johnson avatar
By Todd Johnson
at 2017-05-27T00:22
教主說怕人跳槽啊XD
Eartha avatar
By Eartha
at 2017-05-29T08:24
推文講得好像很嚴重 讓人誤以為zen cpu連gen3都沒
有 誤會大了
Wallis avatar
By Wallis
at 2017-06-01T14:10
ATI那派不是之前被鬥垮惹嗎QQ
應該沒幾個人會想搬到加拿大
Madame avatar
By Madame
at 2017-06-05T05:44
矽谷的人基本上想跑的跑得掉的都跑光惹
上面當官的垮惹 當兵的還在
Liam avatar
By Liam
at 2017-06-05T19:42
但是要包出去怎麼amd沒有licensing 下去嗎
Noah avatar
By Noah
at 2017-06-08T22:39
製程不對 沒辦法QQ
不是某A自製就是另外買QQ
Isabella avatar
By Isabella
at 2017-06-11T03:30
做事又不給子彈(?)
Sandy avatar
By Sandy
at 2017-06-14T18:25
不懂製程不對QQ
Selena avatar
By Selena
at 2017-06-15T17:45
這篇愈說愈替AMD擔心耶,感覺AM4出來後AMD股價該不
會又要QQ一波
Isla avatar
By Isla
at 2017-06-18T02:44
那些ip即使是排前三名的大廠也事會包給你看
Steve avatar
By Steve
at 2017-06-18T05:16
南橋又不是捨摸複雜晶片 沒差喇
Thomas avatar
By Thomas
at 2017-06-18T13:51
不會啦,如果板子也便宜的話XDD
沒有賣不出去的東西,只有賣不出去的價格
Wallis avatar
By Wallis
at 2017-06-19T15:22
到時候amg分別formula g1 xpower 大概還是貴到爆吧
Mia avatar
By Mia
at 2017-06-21T04:33
只要Zen效能有到又便宜 南橋的問題都可以無視
Belly avatar
By Belly
at 2017-06-23T21:46
現在這樣看也只能說不要太慘 效能gg就好了
Robert avatar
By Robert
at 2017-06-27T01:05
結果南橋出大包 SATA原生只有3Gb/s(拖走)
Ivy avatar
By Ivy
at 2017-06-29T00:47
製程對不上大致上就是說ip裡面的設計放到晶片需要的
製程可能會遇到一些限制做不出來
Selena avatar
By Selena
at 2017-07-03T04:02
應該學群暉只有1.5Gb/s (誤
Yedda avatar
By Yedda
at 2017-07-04T23:30
U3S6不要還做到出包就好 其他隨便啦
Quintina avatar
By Quintina
at 2017-07-08T21:36
這倒是不可能啦 AMD 當初也是很早支援6Gbps的
Susan avatar
By Susan
at 2017-07-12T23:09
例如GG28HP跟GG28LP 非常可能不能互通
u3s6便宜 跟pcie2差不多
Genevieve avatar
By Genevieve
at 2017-07-16T20:13
乾脆給VIA做好惹
重返榮耀
Robert avatar
By Robert
at 2017-07-17T04:52
VIA有比AsMedia屌嗎?
Faithe avatar
By Faithe
at 2017-07-19T08:11
笑笑喔 fch 不重要, 你的 cpu lanes 很多嗎
Freda avatar
By Freda
at 2017-07-21T02:05
VIA現在是有是燒紙的狀態
William avatar
By William
at 2017-07-22T01:01
整組phy ip大概百萬鎂左右八
Valerie avatar
By Valerie
at 2017-07-22T18:21
對耶 消費級的zen cpu lane放幾條啊?
Ivy avatar
By Ivy
at 2017-07-23T10:59
8c16t通常應該是16lane的兩組
Candice avatar
By Candice
at 2017-07-24T21:50
16條
Caroline avatar
By Caroline
at 2017-07-25T16:33
apu的可能砍到剩一組半
那兩組的有一部分會被拿去接南僑QQ
Ivy avatar
By Ivy
at 2017-07-28T10:01
4C8T一組這樣?
Hazel avatar
By Hazel
at 2017-07-29T06:10
16好普通的配置
William avatar
By William
at 2017-08-03T03:54
apu 剩 8條
Leila avatar
By Leila
at 2017-08-06T18:11
1組16條嗎,所以這樣2組是32條的意思?
歐我錯了QQ
William avatar
By William
at 2017-08-09T18:42
這.....算悲劇吧QQ
Catherine avatar
By Catherine
at 2017-08-09T19:49
就32條 有一部分會被拿去做別的
Mary avatar
By Mary
at 2017-08-14T02:00
pcie gen3ip其實很吃面積的QQ
而且以前我們team藉口一堆 很敢要die sizeQQ
Emily avatar
By Emily
at 2017-08-15T13:11
ZEN 8c16t 32條比發表會拿出來打的6900k還要少QQ
Anthony avatar
By Anthony
at 2017-08-17T09:53
6900K 有40條
Hedda avatar
By Hedda
at 2017-08-17T13:40
在cpu公司 high speed io地位很高
Anthony avatar
By Anthony
at 2017-08-20T09:20
到惹手機廠就變雜魚天天被幹 反而DDR地位變高惹
William avatar
By William
at 2017-08-20T17:55
手機廠感覺只有水果行有比較注重IO性能
Ula avatar
By Ula
at 2017-08-22T22:49
其他感覺就是可以用就好了,反正大部份消費者無感
Lucy avatar
By Lucy
at 2017-08-26T19:40
iPhone到6s還在用USB2.0耶
Kyle avatar
By Kyle
at 2017-08-30T13:24
手機廠的上頭只管跑分
跑分會看ddr 不會看pcie
Agnes avatar
By Agnes
at 2017-09-01T18:16
還是要看他價錢對應的intel U是幾lane吧QQ
Mary avatar
By Mary
at 2017-09-04T16:28
amd十幾年前屌打intel有一部分就hyper transport
所以io地位高
Puput avatar
By Puput
at 2017-09-06T19:02
zen要是不便宜應該不太好過吧
Isabella avatar
By Isabella
at 2017-09-07T00:12
8c16t應該是可以賣480那葛價錢才對QQ
不過大概不可能QQ
amd說他們server io多也不是唬爛的
四顆8c16t 用gmi連 一顆pcie32條 總共就128條
Jessica avatar
By Jessica
at 2017-09-07T06:16
賣哪個480的價錢?
Olive avatar
By Olive
at 2017-09-09T18:06
跟目前6700k價差不大
Kumar avatar
By Kumar
at 2017-09-13T18:19
現在i社8C16T多少錢啊
Charlotte avatar
By Charlotte
at 2017-09-16T02:46
跟6700K價差不大就農藥了
理想是要i7性能賣i5價
Michael avatar
By Michael
at 2017-09-20T21:33
6900k要一千鎂八
Jacob avatar
By Jacob
at 2017-09-21T01:10
哇靠 查了一下6900K 要37K喔XDDD
Rebecca avatar
By Rebecca
at 2017-09-24T17:51
我覺得4c8t應該是460那葛晶片大小
可是賣100鎂也太便宜惹QQ
Xanthe avatar
By Xanthe
at 2017-09-25T20:32
哇靠33000喔 真是坑爹
Elma avatar
By Elma
at 2017-09-27T17:56
哇那AMD的8C16T價格低於2w就很有趣惹
4C8T賣個199鎂就很殺了吧
Lily avatar
By Lily
at 2017-09-29T05:21
我是覺得8c16t麥葛三百鎂差不多 根i7差不多大小
4c8t應該可以再便宜點
Ethan avatar
By Ethan
at 2017-10-01T05:04
Zen的單核性能都還沒公布,先不要預設能打6900K吧
這篇也說了Zen還有一堆資訊都沒有公布
Suhail Hany avatar
By Suhail Hany
at 2017-10-01T09:50
如果效能真的不輸i7太多
Eden avatar
By Eden
at 2017-10-02T12:49
INTEL賣37K: OK AMD:不賣7K以下沒CP值
Erin avatar
By Erin
at 2017-10-06T11:38
反正還早 QQ
Zenobia avatar
By Zenobia
at 2017-10-06T16:44
樓上上正解
哇那現在intel真的是躺著賺耶
Ingrid avatar
By Ingrid
at 2017-10-10T12:52
還太早了+1
Rosalind avatar
By Rosalind
at 2017-10-10T22:49
教主要出貨AMD前記得要通知一聲啊
Sarah avatar
By Sarah
at 2017-10-15T06:23
可能很多人跟教主已經壓身家進去了QQ
Carol avatar
By Carol
at 2017-10-17T16:29
AMD發表會 ZEN 8C16T 3Ghz 比 6900K 8C16T 3Ghz
強一點
INTEL毛利率過五成,一直都是躺著賺阿
Queena avatar
By Queena
at 2017-10-17T21:27
ZEN出來頂多只能讓他躺的不能超爽
AMD 已經公開講同頻率同核心數能打6900K
Leila avatar
By Leila
at 2017-10-21T05:03
這個看看就好 沒實測出來我不敢相信
Jacky avatar
By Jacky
at 2017-10-22T12:06
那也要ZEN時脈拉的上去
Rosalind avatar
By Rosalind
at 2017-10-26T15:07
那只是跑渲染而已啦 實際一般運用還是個問號
Hardy avatar
By Hardy
at 2017-10-27T14:26
Blender 3D 很有代表性嗎?
Edward Lewis avatar
By Edward Lewis
at 2017-10-29T21:32
如果能在6900K的九成以上就夠AMD邱惹
Andy avatar
By Andy
at 2017-11-03T10:44
問AMD阿 科科
別忘記6900K要價1000USD XD
Tristan Cohan avatar
By Tristan Cohan
at 2017-11-07T14:17
說真的那個跑分應該只是噱頭而已
Carol avatar
By Carol
at 2017-11-10T09:10
若真相教主說,ZEN 8C16T賣300USD 就有競爭力
AMD之前也說RX480 CF打平1080阿 事實上沒說錯
Eartha avatar
By Eartha
at 2017-11-15T02:20
乾 6900K開1000鎂真的是躺著賺
Selena avatar
By Selena
at 2017-11-18T21:37
6900K如果用預設跑跟對比的Zen樣品會不會不相上下
Audriana avatar
By Audriana
at 2017-11-22T18:14
不至於吧 6900K預設有3.7G耶
Isabella avatar
By Isabella
at 2017-11-23T08:46
真正讓intel躺著賺的是伺服器的CPU吧
完全沒對手,只能隨便intel開價
John avatar
By John
at 2017-11-26T03:23
僅限奇點這樣XD
Suhail Hany avatar
By Suhail Hany
at 2017-11-28T10:38
6900K預設時脈3.2Ghz ZEN其他軟體也有八成以上
很強惹
Edwina avatar
By Edwina
at 2017-12-01T02:18
INTEL XEON市佔率不是九成嗎
ZEN的目標就是要在SERVER上重返榮耀 Opteron
Yedda avatar
By Yedda
at 2017-12-01T20:08
消費級 broadwell-e 那幾顆的市場很小眾啦
Callum avatar
By Callum
at 2017-12-04T02:41
樓上不說我都忘記AMD的伺服器CPU叫Opteron惹
David avatar
By David
at 2017-12-04T09:33
就看Naples能不能從E5/E7上搶一些市佔
雙路以上的市場現在真的隨Intel開價 @@
Suhail Hany avatar
By Suhail Hany
at 2017-12-06T05:49
6900K預設時脈3.2Ghz,如果ZEN其他軟體也有八成以上
重點當然是server阿
Heather avatar
By Heather
at 2017-12-10T10:34
那就是 32C64T 的了
Susan avatar
By Susan
at 2017-12-12T16:14
消費級就是看能不能卡住i5 i7中間,讓INTEL跳腳
300USD的市場還是有點距離
Brianna avatar
By Brianna
at 2017-12-14T00:52
怎麼現在又說BDW-E的市場很小眾了0.0
Dora avatar
By Dora
at 2017-12-14T19:09
broadwell-e 本來就很小眾阿 零售有很多?
Ophelia avatar
By Ophelia
at 2017-12-16T01:04
"消費級" Broadwell-E
Cara avatar
By Cara
at 2017-12-17T20:25
話說 am4的 主機板 會跟intel一樣 消費版跟 伺服器
版 不能混用嗎?
哦!
Ingrid avatar
By Ingrid
at 2017-12-21T15:37
AMD伺服器是用SP3 和SP4
不能混用
Jacky avatar
By Jacky
at 2017-12-26T04:54
所以Blender 3D有代表性嗎XD
Sarah avatar
By Sarah
at 2017-12-27T12:09
剛剛不是還說消費級旗艦的FCH規格烙賽 AMD完蛋啦巴
拉巴拉的嗎? 怎麼又說消費旗艦是小眾市場惹
是說我也認同消費級旗艦原本就是小眾
Andrew avatar
By Andrew
at 2017-12-29T22:20
server板不是要用LGA腳座嗎?
Ivy avatar
By Ivy
at 2017-12-30T05:23
X99反而PCH比Z170落後,這點先搞清楚
Ina avatar
By Ina
at 2018-01-03T14:53
太扯
Connor avatar
By Connor
at 2018-01-05T15:39
消費級如果出到8C16T就打到server級市場了
Daph Bay avatar
By Daph Bay
at 2018-01-10T13:23
一般軟體沒有這麼多緒給它,但專業運用
算圖渲染,會有人拿消費旗艦級來用。
Linda avatar
By Linda
at 2018-01-12T09:16
會嗎?
Annie avatar
By Annie
at 2018-01-14T17:51
但是消費級不配ECC RAM會有差吧?
Jake avatar
By Jake
at 2018-01-17T09:18
仍有些專業運用是只追求效能。
Isabella avatar
By Isabella
at 2018-01-18T18:23
應該會有可以配ECC RAM AM4板子吧
Adele avatar
By Adele
at 2018-01-21T01:55
真賣300USD就很吸引人
至少不用花1000USD XD
Anonymous avatar
By Anonymous
at 2018-01-22T19:24
而且ZEN消費級是8C和16C XD
Margaret avatar
By Margaret
at 2018-01-25T16:44
16c32t 和 32c64t AM 4應該就沒辦法 也沒必要了
Olga avatar
By Olga
at 2018-01-26T23:03
可是有些指令集AMD沒有 又QQ
Eartha avatar
By Eartha
at 2018-01-29T09:31
別忘了X99板子的售價.. 整個拉高建構成本
Tom avatar
By Tom
at 2018-01-30T15:25
如果zen的8c16t加上mb賣個13000內 我馬上丟掉intel
Michael avatar
By Michael
at 2018-02-04T14:48
它拿掉一些指令,但也增加不少指令。
Faithe avatar
By Faithe
at 2018-02-06T08:01
哪些沒有? AVX2不也都補齊了?
Erin avatar
By Erin
at 2018-02-07T08:22
我只覺得不用太期待售價,效能夠好,我不覺得會便宜
Jessica avatar
By Jessica
at 2018-02-09T00:37
先前擔心的事情果然發生了
Quintina avatar
By Quintina
at 2018-02-13T05:42
我倒覺得價格可以期待 效能看看就好XD
Mia avatar
By Mia
at 2018-02-13T16:40
把推土機專屬的fma4 xop 之類拿掉吧
Donna avatar
By Donna
at 2018-02-16T04:39
blender那算圖運用就是專業但不在意ECC
Harry avatar
By Harry
at 2018-02-17T04:49
INTEL消費級中階跟高階的空缺是I5~I7還有4C8T以上
AMD只要能做到I5價位I7效能的4C8T
Olga avatar
By Olga
at 2018-02-18T21:51
NVMe我覺得不用急,畢竟是Intel在推的東西。
Kyle avatar
By Kyle
at 2018-02-23T00:12
Anandtech爆料有四顆ZEN在測試的人說 ZEN在AVX和FMA
這兩個指令級上可能會很弱
Christine avatar
By Christine
at 2018-02-26T04:29
還有相同平台就能升到8C16T(想想Z170就能上6C12T!?)
這其實對消費市場都很有魅力
Heather avatar
By Heather
at 2018-03-02T15:48
因為大部分的人都用不到這兩個功能
Yuri avatar
By Yuri
at 2018-03-06T12:47
ACHI一般使用其實夠了,只是SSD現在跑太快。
Erin avatar
By Erin
at 2018-03-10T12:47
A0不是原本就不適合拿來測性能嗎? 上市再說吧
Hazel avatar
By Hazel
at 2018-03-11T00:02
是說X99吃NVME是從CPU接X4就是
Queena avatar
By Queena
at 2018-03-11T07:11
NVMe不用急是沒錯,SSD價格還很高昂
發表會那個應該是一個炒股票的動作
至少證明比K10強多惹
Enid avatar
By Enid
at 2018-03-15T04:59
等到NVMe的裝置取代SATA AHCI再來煩惱啦
Agatha avatar
By Agatha
at 2018-03-18T06:10
反正NVMe可以跟X99、Z97一樣搞追加支援就好
Susan avatar
By Susan
at 2018-03-22T22:01
問題是人家2015丟出來的1XX晶片就有,比起來很落漆
Hedy avatar
By Hedy
at 2018-03-24T00:12
當然會被酸一下丟臉阿
Rosalind avatar
By Rosalind
at 2018-03-28T22:18
AMD本來就落漆阿 何必在意那個XD
Bethany avatar
By Bethany
at 2018-04-02T01:36
也只是比起來 現在買100series晶片組開NVMe的有多
Eartha avatar
By Eartha
at 2018-04-02T10:28
有真的CP值比較重要吧
哪有INTEL有的AMD都要有
Charlotte avatar
By Charlotte
at 2018-04-05T22:33
酸一下沒什麼 只是講的像世界末日一樣有必要嗎XDD
Emily avatar
By Emily
at 2018-04-08T18:14
100系列有些也不支援NVMe,甚至連Z170有些也是追加
Frederic avatar
By Frederic
at 2018-04-11T13:36
這樣AMD會先倒閉十次
Jack avatar
By Jack
at 2018-04-12T20:57
100系列的優勢就是DMI3.0
Quintina avatar
By Quintina
at 2018-04-17T15:23
對呀 講到我以為Promontory還比Bolton還糟咧XD
Olivia avatar
By Olivia
at 2018-04-20T07:48
Z170記得上是20*lane 3.0,然後往上看那個 8*2.0
就會有種很大的落差感覺
Andy avatar
By Andy
at 2018-04-22T03:13
但是要利用到DMI的頻寬優勢,也要有足夠的資料吞吐
Edward Lewis avatar
By Edward Lewis
at 2018-04-22T23:31
唉唷還是那句啦 九成的消費級用戶有U3S6就用不完了
HDD和一些低階SSD還吃不滿S6的頻寬咧
Kelly avatar
By Kelly
at 2018-04-25T11:57
然後這邊一半消費級PC,PCH接不到幾個裝置。
Robert avatar
By Robert
at 2018-04-30T01:34
U3隨身碟是普及不少 但5Gbps還有很大的空間啦
Skylar Davis avatar
By Skylar Davis
at 2018-05-03T18:40
這是感覺問題,要不一堆人買I7也根本買爽的而已
Liam avatar
By Liam
at 2018-05-04T10:16
100系列就好像你家前面開了十線道,可是社區只有兩戶
Bennie avatar
By Bennie
at 2018-05-09T04:29
一般人買i7只是為了時脈吧
Aaliyah avatar
By Aaliyah
at 2018-05-10T07:35
不一定喔 i7是真的日常使用就有感 但這些io可不一定
你還得要花大把的銀兩去買高速儲存裝置才有感覺XD
Brianna avatar
By Brianna
at 2018-05-11T18:36
時脈也要K版才有比較高
Dinah avatar
By Dinah
at 2018-05-15T12:46
其實相容性問題比較重要,或是同一個裝置接不同PCH
的效能表現是否相同。
Necoo avatar
By Necoo
at 2018-05-15T22:23
之前都被酸已經被看到車尾燈 現在能進步到看到車尾
Suhail Hany avatar
By Suhail Hany
at 2018-05-18T02:52
其實對Intel還有一個好處就是行銷上有得說
反正消費者看數字大就是爽,管他用不用得到。
Gilbert avatar
By Gilbert
at 2018-05-23T00:13
燈就很了不起了 先別妄想可以追上
Jacob avatar
By Jacob
at 2018-05-27T17:22
我倒是希望能便宜大碗就好 用不到的東西不重要
Emma avatar
By Emma
at 2018-05-29T15:40
就像機殼上永遠用不到的八大三小。
Dinah avatar
By Dinah
at 2018-05-31T14:10
還是等詳細規格測試出來吧,不過真的沒太差價格不錯
Hedda avatar
By Hedda
at 2018-06-02T12:15
想衝8c16t的
Quanna avatar
By Quanna
at 2018-06-03T14:15
從最近的消息看起來之前爆料4顆ZEN在測試的人應該是
Robert avatar
By Robert
at 2018-06-07T22:36
有所本,單核性能可以期待追上haswell
Lydia avatar
By Lydia
at 2018-06-12T03:19
如果AMD沒搞砸了話QQ
Hardy avatar
By Hardy
at 2018-06-15T22:39
就說拿工程版測的性能都看看就好了。。。
Linda avatar
By Linda
at 2018-06-20T18:44
上市版不會比ES差吧
Damian avatar
By Damian
at 2018-06-22T12:48
先知下次出現不知道什麼時候,想知道更多AMD小秘密
Xanthe avatar
By Xanthe
at 2018-06-27T07:04
我只信教主
Margaret avatar
By Margaret
at 2018-06-28T22:13
ZEN沒有追上haswell那出來幹嘛?丟臉?
Ingrid avatar
By Ingrid
at 2018-07-02T21:00
期望值沒這麼低吧XD
ZEN的設計目標都是拚單核性能了
Andrew avatar
By Andrew
at 2018-07-07T13:57
那時候不是說追上IVB就很欣慰了嗎XD
Agatha avatar
By Agatha
at 2018-07-11T02:50
一開始是說有珊迪就很厲害,現在haswell已經超乎預
期了。更何況AMD還直接拿BDW-E的6900K出來比
Mary avatar
By Mary
at 2018-07-11T18:31
宣傳還拿ivy bridge就真的很漏氣XD
Hamiltion avatar
By Hamiltion
at 2018-07-16T16:31
Intel這幾代單核效能不都是微小幅度提升嗎?
只要追上前幾代,就跟最新的剩下微小差距了。
Daph Bay avatar
By Daph Bay
at 2018-07-18T05:05
不過01的eanck堅稱Zen頂多到SNB就是www
Ingrid avatar
By Ingrid
at 2018-07-18T12:42
當初拿6900k出來對比就已經表示A真的很有野心要拼了
Kumar avatar
By Kumar
at 2018-07-19T02:37
e大濕說他早得知AMD內部資訊www
Eartha avatar
By Eartha
at 2018-07-20T21:50
從SNB的SKL還是有一段不小的提升啦
大概20%有吧
Mason avatar
By Mason
at 2018-07-23T15:32
AMD拿broadwell-e是有所本吧,不然就是想再跌股XD
Oscar avatar
By Oscar
at 2018-07-26T04:04
AMD說ZEN這個架構要戰5年了
不知道ZEN這架構還有多少潛能可以挖
Dorothy avatar
By Dorothy
at 2018-07-26T17:33
這只能說到時候就知道吧
Zanna avatar
By Zanna
at 2018-07-31T16:25
只是不知道這代的浮點運算效能如何
Odelette avatar
By Odelette
at 2018-08-02T20:48
Zen如果改進的方向是對的就沿用阿 跟intel一樣
Elvira avatar
By Elvira
at 2018-08-04T11:53
只會小改不會砍掉重練
Joseph avatar
By Joseph
at 2018-08-08T17:36
i這架構也夠拖了好用確實沒必要換
Faithe avatar
By Faithe
at 2018-08-13T09:08
搞K10的AMD真的像被盜帳號
Ina avatar
By Ina
at 2018-08-17T09:46
教主說intel Core架構的潛能未來會輸Zen
Erin avatar
By Erin
at 2018-08-20T19:54
就內部路線的鬥爭啊,跟當初intel P4差不多
Michael avatar
By Michael
at 2018-08-25T13:05
把Fusion吹成這樣
Jacob avatar
By Jacob
at 2018-08-27T03:38
只是intel底子夠厚,AMD一次架構大失誤就QQ了
Tom avatar
By Tom
at 2018-08-31T06:44
netburst真的頗強,版上大濕還說他很穩定
行銷一級棒
Skylar DavisLinda avatar
By Skylar DavisLinda
at 2018-08-31T09:08
Fusion的確才是未來 但他確實也還在未來
Ula avatar
By Ula
at 2018-09-04T02:53
只是AMD有那個本錢可以CPU設計路線鬥爭嗎
Anthony avatar
By Anthony
at 2018-09-05T22:43
P4行銷超強大,intel inside深植人心
Sarah avatar
By Sarah
at 2018-09-08T20:23
應該是提升的潛能啦 好比INTEL已經90分了 空間有限
zen可能還在75分而已 畢竟剛新設計的東西 還有空間
Gilbert avatar
By Gilbert
at 2018-09-10T08:56
反正大概跟顯卡一樣 初期的產品一定還是賣輸對手
Charlotte avatar
By Charlotte
at 2018-09-12T20:57
畢竟對手的東西也不差 而且更深植人心
Necoo avatar
By Necoo
at 2018-09-13T05:50
我最大的期望就是6K有AMD的CPU可以選
Lucy avatar
By Lucy
at 2018-09-16T03:35
http://tinyurl.com/gs83tg7 參考一下時代的眼淚 XD
Elizabeth avatar
By Elizabeth
at 2018-09-16T23:04
最強的Pentium_D_960 = E2160
Elizabeth avatar
By Elizabeth
at 2018-09-17T19:37
最強的PD不是EE版的嗎XD 2C4T耶
Agatha avatar
By Agatha
at 2018-09-22T11:53
http://tinyurl.com/jgdf7en 好像差不多... =.= XD
Lucy avatar
By Lucy
at 2018-09-27T09:17
A的high spedd io整團就是從V過去的啊...抖抖
Megan avatar
By Megan
at 2018-09-28T12:00
PE 965效能跟Celeron G470u一樣
PE 965效能跟Celeron G470一樣
Andy avatar
By Andy
at 2018-09-28T23:55
從威盛跳到祥碩嗎QQ
Queena avatar
By Queena
at 2018-10-01T17:14
還有AMD Athlon 64 X2 4200+ (2.2Ghz)
Celeron G470是2Ghz 時代眼淚
John avatar
By John
at 2018-10-05T11:15
anandtech 上面還有一篇看伺服器主機板的也不錯
Kristin avatar
By Kristin
at 2018-10-09T15:39
如果不是asm還能找誰代打?三個字母的那票四五年前就
是空殼了..
Sierra Rose avatar
By Sierra Rose
at 2018-10-13T09:17
找marvell上嗎?
Barb Cronin avatar
By Barb Cronin
at 2018-10-17T12:34
eanck也是很會吹 算了吧
Poppy avatar
By Poppy
at 2018-10-21T09:18
另外重點還是在於你都打算16/17推出了了
端出來的的spec還是這種兩光
Quintina avatar
By Quintina
at 2018-10-22T05:36
nvme 只是裝pcie2.0會碰到的一個瓶頸而已
Bethany avatar
By Bethany
at 2018-10-25T15:52
ASM挖威盛的的人過去之前不是還有上新聞XDDD
Mia avatar
By Mia
at 2018-10-27T07:58
是說VIA現在還剩什麼阿,好歹他也是x86第三大廠(?
當初EPC流行的時後也出過幾台VIAx86的機器
Poppy avatar
By Poppy
at 2018-10-28T19:35
先搶爆
有ssd控制器XDDDDDDDDDD
Michael avatar
By Michael
at 2018-10-29T08:02
做中國的cpu,沒意外今年底明年初tapeout,還有一些
project
Edward Lewis avatar
By Edward Lewis
at 2018-11-01T00:17
Via好歹有寶貴的x86授權。太浪費了。
Ula avatar
By Ula
at 2018-11-02T06:15
拿去跟中國共用了啦,叫Viacpu
Kyle avatar
By Kyle
at 2018-11-03T03:04
內建祖國善意指令集
Bethany avatar
By Bethany
at 2018-11-07T16:12
eanck就比u大師還寶啊 各種唬爛不打草稿
Mason avatar
By Mason
at 2018-11-09T15:17
去年有開16nm的testchip,不知道進度如何,應該是沒
救啦
Noah avatar
By Noah
at 2018-11-11T03:09
還說SATA轉接線燒了 把線剪開焊fuse比較保險www
整天在01上說自己是AMD股東 可以調到機密資料www
Callum avatar
By Callum
at 2018-11-15T23:30
結果整天只能在01上靠北AMD 不知道有沒有本事影響
AMD公司策略
George avatar
By George
at 2018-11-16T10:38
銲保險絲是認真的嗎wwww
Frederic avatar
By Frederic
at 2018-11-16T12:55
中國芯 玻璃芯
Oliver avatar
By Oliver
at 2018-11-19T09:59
可是光是AMD股東那句就是在嘴的吧w
Harry avatar
By Harry
at 2018-11-23T14:16
http://goo.gl/LvjeBM
想用molex to sata 轉接線? 先焊上"保險絲"才能多一
層保障。
Franklin avatar
By Franklin
at 2018-11-28T11:45
VIA現在的價值就剩x86授權了吧,S3的部分賣給紅茶了
Andrew avatar
By Andrew
at 2018-12-01T11:05
買一張股票也算股東啊
Iris avatar
By Iris
at 2018-12-02T18:57
一股也算股東啊XDDDDDDD
Lily avatar
By Lily
at 2018-12-06T18:53
VIA的x86連對岸都看不太上了
Barb Cronin avatar
By Barb Cronin
at 2018-12-08T05:33
eanck早就被壞蛋酸過了 各種奇文共賞
Hazel avatar
By Hazel
at 2018-12-10T10:25
看VIA的網頁現在好像都在作一些嵌入式的晶片
Andy avatar
By Andy
at 2018-12-13T17:11
對岸拿錢跑去跟AMD合作。VIA哭哭
Xanthe avatar
By Xanthe
at 2018-12-18T15:27
http://goo.gl/zNpgPi eanck暢談AMD營運策略
Skylar Davis avatar
By Skylar Davis
at 2018-12-23T04:16
VIA簽的x86授權是不是跟AMD一樣轉賣就失效了?
Suhail Hany avatar
By Suhail Hany
at 2018-12-25T09:22
要當AMD股東還不簡單XDD
Rachel avatar
By Rachel
at 2018-12-28T23:51
根本奇文共賞耶 還是那個chrislc腦袋比較清楚w
Mia avatar
By Mia
at 2019-01-02T11:42
焊保險絲那太神了居然還有人信www
Linda avatar
By Linda
at 2019-01-06T15:13
最怕的是有人信了結果焊爛cold joint
Ethan avatar
By Ethan
at 2019-01-08T19:59
我看了半天才覺得為什麼這焊保險絲跟我所知的短路保
護不一樣呢
William avatar
By William
at 2019-01-09T19:00
6年前分析師覺得NV應該要買Via
後來NV選擇搞Arm往手持。
Frederica avatar
By Frederica
at 2019-01-14T18:39
eanck暢談AMD營運策略XDD
Olive avatar
By Olive
at 2019-01-18T01:23
....保險絲....!!
Rachel avatar
By Rachel
at 2019-01-18T03:19
問他股票買在多少 我原價跟他買讓他無痛停損
Daniel avatar
By Daniel
at 2019-01-20T03:50
不問鄉民知多少 只問股票買多少
Isabella avatar
By Isabella
at 2019-01-22T06:07
一整串看下來有點累
反正就快點出來,性價比好看就會有人買單了。
Emma avatar
By Emma
at 2019-01-22T12:03
Via買下去vn會QQ嗎
Donna avatar
By Donna
at 2019-01-26T06:52
能讓兩邊打到難分難捨最好,爽到消費者撿便宜www
Daph Bay avatar
By Daph Bay
at 2019-01-28T04:10
教主只有20nm (過錫爐)
Dora avatar
By Dora
at 2019-01-31T06:03
桌上型賺不到大錢啦 炒股票用的QQ
Hedda avatar
By Hedda
at 2019-02-02T14:00
eanck確定不是大濕在01的帳號嗎XD
Heather avatar
By Heather
at 2019-02-06T07:37
我懷疑wow才是他的帳號,常常捧那文章
Noah avatar
By Noah
at 2019-02-09T13:58
wow是大濕的恩濕好嗎 是濕父
Gilbert avatar
By Gilbert
at 2019-02-11T05:53
轉接線焊上保險絲............他講的好認真
Tristan Cohan avatar
By Tristan Cohan
at 2019-02-12T13:16
會燒基本上是線材的問題吧,怎麼會跑去上保險絲= =
有圖嗎
George avatar
By George
at 2019-02-15T20:28
blender 天天用啊,雖然我都開 GPU 加速了.....
Lucy avatar
By Lucy
at 2019-02-19T05:51
E5 兩顆也比不上一張 R9 380
Todd Johnson avatar
By Todd Johnson
at 2019-02-23T16:36
eanck還有出現在pcdvd過 還唬爛到被壞蛋洗臉咧
Rebecca avatar
By Rebecca
at 2019-02-28T01:27
不過貼圖超大一次吃掉幾十G ram 的時候就只好看 CPU
Kelly avatar
By Kelly
at 2019-03-02T18:52
所以ZEN 8C賣300USD對你有搞頭嗎
Odelette avatar
By Odelette
at 2019-03-03T21:23
vobor有錢買人家股票沒錢買10張卡?....
Dorothy avatar
By Dorothy
at 2019-03-04T02:21
大濕跟西斯教派一樣採師徒制喔?
Hedwig avatar
By Hedwig
at 2019-03-05T14:52
這篇怎麼沒看到大濕出來呢XD
Charlotte avatar
By Charlotte
at 2019-03-08T16:09
CEO蘇媽有一百萬股 幾葛SVP也都五十萬股等級
好爽的說QQ
Olive avatar
By Olive
at 2019-03-12T16:21
今天AMD沒有繼續往上衝 我難過QQ
Enid avatar
By Enid
at 2019-03-13T14:29
明天晚上hotchip AMD就要講惹 兩天後再看看八QQ
Charlotte avatar
By Charlotte
at 2019-03-18T13:03
我也在等hotchip 感覺AMD應該會講ZEN的新消息
Doris avatar
By Doris
at 2019-03-19T03:16
今日無往上才是對的
Ethan avatar
By Ethan
at 2019-03-19T11:49
我只是在盤前買了幾股覺得今天跌了有點難過QQ
Yuri avatar
By Yuri
at 2019-03-22T07:16
不過為何說今天沒往上才是對的
Rosalind avatar
By Rosalind
at 2019-03-26T17:19
東西都還出來看到一根毛是在漲殺小
Hamiltion avatar
By Hamiltion
at 2019-03-27T10:01
Zen真的好 漲到70也小菜一疊
Eden avatar
By Eden
at 2019-03-29T13:29
下一季財報EPS轉正 才會穩定往上漲 不是靠打嘴砲
Ingrid avatar
By Ingrid
at 2019-04-01T12:58
終於全看完了!
Edwina avatar
By Edwina
at 2019-04-02T15:23
這篇太神辣
Daph Bay avatar
By Daph Bay
at 2019-04-06T13:30
SATA轉接線焊上保險絲?!認真的嗎?囧

25K打lol、文書、遊戲機(含OS)

Xanthe avatar
By Xanthe
at 2016-08-22T12:58
已買/未買/已付訂金(元): 預算/用途:表弟上大學,說平常遊戲只有打lol,預算25k。 CPU (中央處理器): Core i5-6500 MB (主機板): 華碩 B150M-A/M.2 RAM (記憶體): 美光 DDR4-2133-8GB VGA (顯示卡): 微星 R ...

PCIe4.0明年初發佈頻寬翻番5.0已提上日程

Zanna avatar
By Zanna
at 2016-08-22T12:10
[摘要]PCIe 4.0不但頻寬翻番,它能耗還更低,這無疑將拓寬其應用範圍 舊版相容性使得它能支援以往的設備。此外,雖然PCIe 4.0規範要到明年初才能發佈,但 5.0的開發已經提上議事日程。 據PCMag 網站報導,PCI Express(PCIe)匯流排標準速度已經相當快,功能也已經相當強 大。但它可能 ...

Intel Kaby Lake 將會帶來最強的14nm晶片

Genevieve avatar
By Genevieve
at 2016-08-22T12:09
去年,桌面晶片巨頭英特爾發佈了自家極具影響力的 Skylake 微架構處理器 雖然已經是 14 納米工藝的第二年時間裡,但卻代愈發成熟,更適用於高性能的桌面平臺 難能可貴的是,英特爾終於不再只是鼓吹針對筆記型電腦或平板電腦的低端酷睿處理器有 多牛 而是回歸到了所熟悉的桌面領域。 為此,英特爾花了很長時間,特 ...

EVGA發布GTX1060 ACX3雙風扇SSC/FTW FTW+

Mason avatar
By Mason
at 2016-08-22T12:09
EVGA 在官網發布了 OC 增強版 以及採用 AC 3.0 雙風扇版本的 GTX1060 6GB SSC http://www.evga.com/Products/Product.aspx?pn=06G-P4-6264-KR FTW http://www.evga.com/Products/Produc ...

慶祝 30 週年 msi 準備推出GTX1080 30th

Gilbert avatar
By Gilbert
at 2016-08-22T12:08
CPU散熱器搭配這個神秘的顯示卡目前在德國的GamesCom展示。 去年MSI為了慶祝破億張GeForce顯示卡銷售推出了特別版的顯示卡 今年微星為了慶祝上市30年。這對台灣公司是一個很大的週年紀念 所以他們不得不準備一些特別的東西,像一個新的高階顯示卡。 GeForce GTX 1080 30th A ...