PCDVD數位科技討論區
PCDVD數位科技討論區   註冊 常見問題 標記討論區為已讀

回到   PCDVD數位科技討論區 > 電腦硬體討論群組 > 顯示卡討論區
帳戶
密碼
 

  回應
 
主題工具
Adsmt
Golden Member
 
Adsmt的大頭照
 

加入日期: Feb 2004
您的住址: 從來處來
文章: 2,765
這下換 ATi 了, ATi 顯卡傳瑕疵

http://computershopper.com/shoptalk...hip-with-defect


Some AMD ATI Radeon HD 4830-Based Graphics Cards Ship With Defect

You'd think that in light of the recent fiasco Nvidia endured with a large swath of one of its GPU families being found defective and related products recalled by their respective manufacturers, that AMD might learn a valuable lesson from Nvidia's mistakes and not venture where Nvidia had just tread. But, as they say, history is apt to repeat itself.

No sooner had AMD sent graphics card reference designs using its new ATI Radeon HD 4830 GPU to review sites, that something was not quite right with some of the cards started bubbling up to the blogosphere surface. When the folks over at bit-tech.net tested their Radeon HD 4830 card they were getting lower-than-expected benchmark scores that didn't make sense--especially when compared against the Radeon HD 4850. They also saw similarly bizarre performance results from an HIS-supplied graphics card also using a Radeon HD 4830 GPU. They weren't the only ones to see this issue...

When the techPowerUp tester got his hands on the ATI Radeon HD 4830 reference card he was gathering technical specifications of the card using the GPU-Z utility when he noticed that the utility was seeing only 560 stream processors. The Radeon HD 4830 is supposed to have 640 stream processors--80 stream processors were missing. This means that Radeon HD 4830 cards with 560 stream processors could potentially perform as much as 10 percent slower than cards with 640 stream processors.

It turns out that AMD mistakenly sent out reference-design cards to the media with the wrong BIOS installed. In addition to this being egg on AMD's face, a number of the Radeon HD 4830 product reviews that have been posted show performance based on the presumably wrong, under-performing BIOS. Making matters worse, AMD reports that a small number of Radeon HD 4830 boards from HIS with this same wrong BIOS version are actually in the channel now, where consumers could potentially buy them. The issue is fixable with a BIOS update, but the question is, will everyone who purchases one of the impacted products learn that they need to manually update the card's BIOS, and will they even want to go through the hassle?

Here is ATI's official response to bit-tech.net on the issue:

"AMD has identified that, in addition to reference samples of the ATI Radeon™ HD 4830 boards sent to media with a pre-production BIOS potentially impacting the card’s performance, a very limited number of ATI Radeon™ HD 4830 boards were released to market with the same pre-production BIOS. This is in no way hardware related, and an updated BIOS fully resolves the performance limitation.

Through consultations with AMD board partners, it has been determined with a high degree of certainty that fewer than 400 ATI Radeon™ HD 4830 boards from one AMD board partner, HIS, have reached the market with the pre-production BIOS incorrectly provided by AMD. As only a small number of HIS-branded ATI Radeon™ HD 4830 cards are impacted, we ask any customers that purchased an HIS-branded ATI Radeon™ HD 4830 to test the board using the GPU-Z utility (available at http://www.techpowerup.com/gpuz). If the GPU-Z utility reports fewer than 640 shaders, please visit the HIS website for information on how to update the card BIOS via a downloadable install utility."

大義:
有部份的 4830 的顯卡,大約 10% SP 無作用,ATi 官方回應是因為 BIOS 的問題,更新 BIOS 就沒事了,而且僅有少部份的 4830 顯卡有問題。

這些回應怎麼似曾相似.....
     
      
舊 2008-10-28, 09:57 AM #1
回應時引用此文章
Adsmt現在在線上  
linczs2000
Master Member
 
linczs2000的大頭照
 

加入日期: Mar 2002
您的住址: 屏東<===>台北
文章: 1,578
這問題我們在台灣應該不需要擔心

因為我們能入手的都是非公版的4830

現在連4850要找到公版貨都不大容易
 
舊 2008-10-28, 10:11 AM #2
回應時引用此文章
linczs2000離線中  
Boyoungirl
Power Member
 
Boyoungirl的大頭照
 

加入日期: Aug 2004
您的住址: EARTH
文章: 691
免驚免驚...
現在A家氣勢旺

可以開啟 無視大法
__________________
旅行於陌生之境,意義總在終點才會浮現。可重回熟悉之處,卻步步重拾了記憶,處處彰顯了言外之意。

舊 2008-10-28, 10:20 AM #3
回應時引用此文章
Boyoungirl離線中  
ikari
Junior Member
 
ikari的大頭照
 

加入日期: Oct 2000
文章: 720
印象中公版560SP的4830只出貨不到四百張
而且透過BIOS即可更新為640SP
所以毋須擔心

而且就像樓上所言...要在國內買到公版4830簡直難如登天,好像只有HIS有出貨

比起NV那種瑕疵算是小巫見大巫
舊 2008-10-28, 10:37 AM #4
回應時引用此文章
ikari離線中  
caprichilde
Major Member
 
caprichilde的大頭照
 

加入日期: Apr 2007
您的住址: 台北趴場
文章: 213
引用:
作者ikari
印象中公版560SP的4830只出貨不到四百張
而且透過BIOS即可更新為640SP
所以毋須擔心

而且就像樓上所言...要在國內買到公版4830簡直難如登天,好像只有HIS有出貨

比起NV那種瑕疵算是小巫見大巫

標題殺人?
話說的太快了,兩種問題差距太遠。
一個是製造材料上出了問題,一個是BIOS問題,
一邊是遍及全部產品線,一邊只有公版4830有問題。

然後我朋友還呆呆的買了8600GT的NB 我只說那你好好使用,用到他壽終正寢 。
舊 2008-10-28, 12:58 PM #5
回應時引用此文章
caprichilde離線中  
Boyoungirl
Power Member
 
Boyoungirl的大頭照
 

加入日期: Aug 2004
您的住址: EARTH
文章: 691
8600GT...買來了也算是產品瑕疵

不能直接無痛升級嗎
__________________
旅行於陌生之境,意義總在終點才會浮現。可重回熟悉之處,卻步步重拾了記憶,處處彰顯了言外之意。

舊 2008-10-28, 01:33 PM #6
回應時引用此文章
Boyoungirl離線中  
tch68818
Power Member
 

加入日期: Nov 2003
文章: 646
BIOS有問題=晶片全線有問題?

樓主發這篇是來當笑話的?


或許小的一直都在用ATI晶片的顯示卡
到現在也沒聽過ATI的晶片有設計瑕疵
不像NVIDIA已經賣了一大堆才發現晶片有瑕疵
竟然還有改過編號的問題晶片在外面流

賣不完改晶片編號大清倉可以理解
已經抓到有問題還讓改晶片編號的問題晶片在外面流是怎樣?
__________________
工作機
Ryzen 5 7600
AsRock B650m Pro RS
Klevv DDR5 6000 16G x2
Team G70 2T NVME SSD
Toshiba 4T / 7200 RPM
PowerColor RX6650XT
FSP GT Pro 1000W
舊 2008-10-28, 02:00 PM #7
回應時引用此文章
tch68818離線中  
willism
Master Member
 
willism的大頭照
 

加入日期: Mar 2006
文章: 1,772
這消息是所謂的瑕疵呢,還是只要刷bios就可以開管(可以直升HD4850 ?),

這兩點的差異非常大喔
舊 2008-10-28, 02:06 PM #8
回應時引用此文章
willism離線中  
weredc911
*停權中*
 

加入日期: Feb 2007
文章: 225
真是缺德的"標題殺人法"
看都看無還大義...我還大XX
舊 2008-10-28, 03:32 PM #9
回應時引用此文章
weredc911離線中  
惡魔印記666
Master Member
 

加入日期: Jul 2004
文章: 1,854
這消息早在上星期就有了

ATI是因為公版出貨時
BIOS閹割的太多才出問題
刷個新BIOS就解決了

NV出的包可不是刷個BIOS就可以的

兩個嚴重程度天差地別
一看就知道N飯出來搞標題殺人法

今天中午才遇到有個朋友MSN問這個消息
他也是差一點就被標題誤導.....
此風不可長
__________________
玩硬體就像是瞌藥一樣
很容易上癮.....胃口也漸漸變大
想撩進去的人要三思啊 orz
舊 2008-10-28, 04:08 PM #10
回應時引用此文章
惡魔印記666離線中  


    回應


POPIN
主題工具

發表文章規則
不可以發起新主題
不可以回應主題
不可以上傳附加檔案
不可以編輯您的文章

vB 代碼打開
[IMG]代碼打開
HTML代碼關閉



所有的時間均為GMT +8。 現在的時間是12:59 AM.


vBulletin Version 3.0.1
powered_by_vbulletin 2025。