Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
The Gigabyte GA-MA790FX-DQ6 has six USB ports on the backplane and 2 eSATA ports. The confusion may come from the fact that two of the USB ports are orange like the eSATA ports are instead of black. They are also right above the eSATA ports.
Any word on when the first 790FX (or any of its derivatives) reviews will be up?
The review is already written, but I have no idea when it will go up. My guess is that it won't happen until after Christmas.
Send it to me and I will help you host it somewhere
No, since Sandra's mistake in measuring memory bandwidth has no bearing on other application performance tests.In reference to benchmarks...
Do you think there'll be a need to re-evaluate the programs we use to measure performance, and the rapidity with which we use them to throw up the fastest review of a new product? What do you think about the change in benchmark numbers?
No, since Sandra's mistake in measuring memory bandwidth has no bearing on other application performance tests.
Kyle: You asked about who had been saying the TLB caused issues with everyday stuff, especially CS3 - it was ExtremeTech who said it, but from the quick look I gave the article, it looks like they were talking about After Effects CS3, not PS CS3.
I've also seen mention [only a forum post so far] of an alleged issue with 3dMark 2006, but have yet to find any actual articles mentioning it.
It wouldn't surprise me if people are blaming crashes from other things (messed up windows install, poor drivers, etc...) on the TLB errata...
Maybe. I can't test it myself as a) I don't have a Phenom, & b) other than the Flash player plugin, I don't have any Adobe software.
sourceArstechnica said:The company is also confident that end-users will never experience the TLB bugso much so, in fact, that it intends to offer end-users the chance to turn the required BIOS fix on and off at will. Future versions of AMD Overdrive will include such a toggle, thus allowing enthusiasts to choose for themselves which mode to run in.
I tried to explain the non-effect of the TLB issue in a Phenom thread on Slickdeals. I think some people called me a n00b and Kyle a hack.
The TLB problem honestly shouldn't be there in the final silicon. AMD screwed up and it is as simple as that. The fanboys would have you believe that it is a totally minor issue that is totally unimportant. For some people the problem won't ever be an issue but in some scenarios it clearly will be. Still the fix cripples the performance of the processor and the TLB problem is just one more reason why the Phenom is less desirable than the Core 2 Quad processors are.
Yes- it was a mistake, NEVER should have been released without fix.. it happens.
Fanboys?
MOST folks will never encounter this issue, not SOME people.
There are SOME people that will encounter this issue(they've stated it's a VM issue- MOST folks DON'T utilize that), but the % is SMALL is it not?
My WAG... they will drop.
Internal disagreement?
This is not directed toward anyone in particular, but more as a general statement: A new chip takes several years to develop, and 70% of that time is spent testing, verifying, and validating. Literally hundreds of millions of test hours have been spent in various forms testing the chip. The fact that it was caught so late in the process shows just how extremely rare it is. Should it have been caught earlier? Definitely. But things like this occasionally happen when you are dealing with permutations between several hundred million elements. It sucks, but c'est la vie. Should it be fixed? Most definitely, and it will be very soon.
Are you considering going with Phenom in the future morfinx ?
In reference to benchmarks...
Kyle,
Nathan Kirsch over at Legit Reviews has a few articles up: the first is echoing what you said, that the whole TLB issue is a fart in the wind. The second is very interesting in that apparently SiSoft Sandra XII wasn't properly measuring Phenom bandwidth... reflecting a decrement of HALF the actual bandwidth. I did as he did and downloaded SiSoftware's SandraXII SP1 and my bandwidth score jumped from 5700+ to 10.6G! That explains why I was getting great memory marks in Eerest Home while Sandra was reporting screwed pooch scores.
Do you think there'll be a need to re-evaluate the programs we use to measure performance, and the rapidity with which we use them to throw up the fastest review of a new product? What do you think about the change in benchmark numbers?
By using the 'turbo' button one can disable the TLB BIOS patch when using the yellow or red 'turbo' settings. By pressing the turbo button the ring around the button will change color as seen in the picture above. We asked AMD what the different settings on the turbo button do and we were informed the following.
Green = safe
Yellow = faster
Red = fastest
It doesn't adjust any clock speeds.
We asked AMD if they could go into more details on what exactly these settings are doing, but it has been days and they have not commented. To be honest we have asked more than one person in AMD what these settings do and either they don't know or don't reply.
I will definitely buy this. But i will most likely wait till B3s are out or some kind of price drop to bring it closer to $200.
How can someone call morfinx a n00b? That is so wrong.
Morfinx, how heavy are your discounts buying from AMD? LOL? Do you get 2-for-1 deals on older procs?
Where some websites have dealt FUD
The bottom line is that the websites making a big deal over this to bring in the page hits have yet to be able to produce the errata error in a desktop environment.
When we updated the BIOS to the most recent one, version 0603, released on the ASUS Web site, we were able to get After Effects CS3 to run and complete the benchmark. The end result is pretty slow, compared to the Q6600. But the stability problem we initially encountered looks to be a thing of the past.
And you have to remember, too, that the Phenom parts that we launched were really targeted at the mainstream computer user and not the enthusiast guys. So the mainstream computer user really isn't going to notice any impact.
Demski said that the errata bug affects highly threaded applications due to a problem in translation between the L2 and L3 cache on the chip, which resulted in data being overwritten.
Saucier confirmed that the TLB erratum can cause the system to hang when the chip is experiencing high utilization. AMD has stated previously that virtualization workloads can lead to this problem, but Saucier clarified that other workloads can trigger system hangs, as well.
Which? All CPUs have "bugs", but only some are deemed serious enough to recall, stop production, be patched. What's FUD in reporting it?...
If you're a real cheapskate though, that $199 2.2GHz 9500 doesn't look too bad.
You are picking up a BE, yes?
Yes,In fact I have already ordered one locally.
"We haven't changed the shipping pattern," AMD man Phil Hughes told InternetNews. "It's only a stop ship if it's shipping in volume, and we're only shipping Barcelona for specific customer commitments, like larger volume deployments."
AMD seems to be fiddling with language, as far as we're concerned.
For one, we're told that this isn't a "stop ship" because as far as AMD is concerned it's not really shipping the products. Erm, ok.
Oh yea screw Intel.
Why do you hate Intel so much? Right now they have the better performing parts and those can be had in just about every price point.
Well you have to admit there is a more than a sneaking suspicion that they have muscled the market/industry and by doing that sold more chips. Imagine if AMD had caught a break like Dell or something earlier? R&D funds (or funds period) might not be an issue. Know what I'm saying?
Paying anyone to look the other way or give you an advantage over the other guy is a bribe. It's just that simple. I'm sure it's happened before, and still does. Still doesn't make it right.I still don't get it. Business is business. Intel is basically doing what other companies used to do with Comp USA. Basically some companies would pay Comp USA to not sell other companie's products. This wasn't a secret and it wasn't illegal as far as I know. If anything, Intel might be doing something similar. It may be "unfair" but the fact is the world isn't fair.
For the most part Intel has been more profitable because they've had better products for most of the last 20 years. From 1999 or 2000 onward AMD was far more competitive than they had ever been in the past, but they just haven't had the ability to beat Intel on a business level. That was a situation where having the better product is only part of the battle.
Why do you hate Intel so much? Right now they have the better performing parts and those can be had in just about every price point.
I like Intels Products and if I could stand the company I might buy more. But with personal dealings I have had with them over the years, I perfer to buy AMDs products even if they they dont perform as well as Intels. Its one of those vote with your money things
I buy the best products I can afford. I don't really give any thought to the company's politics or behavior.
d. Threats of Retaliation
72. Beyond exclusive dealing, product and channel restrictions and exclusionary rebates, Intel has resorted to old-fashioned threats, intimidation and knee-capping to deter OEMs from dealing with AMD. Intel has a variety of pressure points at its disposal: it can unilaterally reduce or withdraw a discount, rebate or subsidy; it can impose a discriminatory price increase on a disfavored customer, extend a price cut to that customers competitor, or force retailers into dropping the customers computers and buying from its competitor instead; or it can delay or dispute an allowance or rebate all of which can turn a profitable quarter for an OEM into an unprofitable one. Other pressure points on accounts it deems disloyal include threatening to delay or curtail supplies of scarce processors or essential technical information. Examples abound.
73. As Gateway executives have recounted, Intels threats beat them into guacamole. But Gateway is not alone. Prior to its merger with HP, Compaq Computer received Intel threats every time it engaged with AMD. In late 2000, for example, Compaqs CEO, Michael Capellas, disclosed that because of the volume of business he had given to AMD, Intel withheld delivery of server chips that Compaq desperately needed. Reporting that he had a gun to his head, Capellas informed an AMD executive that he had to stop buying AMD processors.
74. In 2002, Intel pointed its gun at NEC. Intel threatened to discontinue providing NEC with the technological roadmap of future Intel products if NEC did not convert its entire line of Value Star L computers to Intel microprocessors. Without that roadmap, NEC would be at a distinct competitive disadvantage. Predictably, NEC succumbed and eliminated AMD from the Value Star L series in 2002 and 2003.
75. NECs European subsidiary, NEC-CI, which operates NECs European and non-Japanese Asian divisions, reported that Intel executives said they would destroy NEC-CI for engaging with AMD in the commercial desktop segment. Intel told NEC-CIs retailers that NEC-CIs AMD dealings could impair its ability to supply products to its customers, and when NEC-CI resisted the pressure, Intel imposed a discriminatory price increase.
76. AMD had been engaged in discussions with IBM about introducing an Opteron blade server, when IBM suddenly announced that any such product it distributed could not bear an IBM logo. When pressed for an explanation, IBM reported that it could not appear overly supportive of AMD server products because it feared Intel retaliation.
e. Interference with AMD Product Launches
77. Key to gaining quick market acceptance of a new microprocessor is a chipmakers ability to develop a lineup of reputable launch partners, consisting of OEMs prepared to roll out products featuring the chip, major customers who are willing to buy and embrace it, and other industry allies, such as major software vendors and infrastructure partners who can attest to its quality and reliability. Particularly for commercial and enterprise (i.e., server-work station) purchasers, a successful and impressive launch is essential to generating confidence among the computer professionals who will be the potential audience for the new microprocessor.
78. Aware of the importance of product launches, Intel has done its utmost to undermine AMDs. Set forth below are several examples.
79. AMDs September 23, 2003, launch of Athlon64 was a watershed event for the Company. Upon learning the launch schedule, Intel did its best to disrupt it. For example, Acer committed to support the AMD rollout by making a senior executive available for a videotaped endorsement and by timing the introduction of two computers, a desktop and a notebook, to coincide with AMD events planned for Cannes, San Francisco and Taiwan. Days before the event, Intel CEO, Craig Barrett, visited Acers Chairman, CEO and President in Taiwan, expressed to them Intels concern and said Acer would suffer severe consequences if it publicly supported AMDs launch. The Barrett visit coincided with an unexplained delay by Intel providing $15-20 million in market development funds owed to Acer. As a result, Acer withdrew from the launch in the U.S. and Taiwan, pulled its promotional materials, banned AMDs use of the video, and delayed the announcement of its Athlon64-powered computers. Acers President subsequently reported that the only thing different about Intels threats was the messenger they were usually done by lower ranking managers, not Intels CEO.
80. HP also withdrew precipitously from the Athlon64 launch after committing to participate. HP had agreed to support the launch by producing a promotional video and by sending senior executives to all three launch sites. Just before launch, however, HP manager, John Romano, pulled the video and announced that HP would only be sending a junior manager, and then only to Europe.
81. Other AMD customers and channel partners reporting Intel coercion to withdraw from the Athlon64 launch were Lenovo, NEC-CI and Best Buy.
82. Intel also disrupted AMDs launch of its Opteron server chip, which was rolled out on April 22, 2003, with few in attendance and little industry support. A computer industry journal reported Intels fingerprints: They all [vendors] told me that prior to the launch, they received a phone call from Intel. Intel asked if they were going to the launch. If they replied yes, the Intel rep asked them if it was important to them to go, or if they really wanted to go. Pressing the vendors, I got the same response, Intel is too smart to threaten us directly, but it was quite clear from that phone call that we would be risking our various kickback money if we went.
83. Other companies that reported being intimidated from participating in the Opteron launch were MSI, Atipa, Solectron and Fujitsu-Siemens. Indeed, Intel representatives told Fujitsu-Siemens executives in the weeks preceding the Opteron launch that if they attended, they would be the only Tier One OEM showing its support as all of the others would back out. With the exception of IBM, Intel was right.
84. These are not isolated examples, but rather illustrations of Intels relentless campaign to undermine marketing efforts by its one remaining competitor. For example, IBM pulled its AMD-powered computers from the 2004 Palisades eServer and PC Show, citing a contractual agreement with Intel said to prohibit it from endorsing those competitive products. And at the 2004 Super Computing Show, an annual conference devoted to high performance computing, Intel offered two other AMD customers money to remove AMD systems from their booths. At CeBit, Intel threatened to pull a half million dollars of support from Fujitsu-Siemens for displaying AMD products (which were removed).
There is absolutely no problem with buying a product based on its perceived value. But people should be aware of some of Intel's past behaviors, as outlined in the full complaint filed by AMD in Delaware. I encourage everyone to read it in its entirety (particularly page 15-41), as the below is just a small fraction of what's in the complaint:
http://www.amd.com/us-en/assets/content_type/DownloadableAssets/AMD-Intel_Full_Complaint.pdf