Friday, February 13th 2009

AMD Clarifies Erratum 379 ''DDR3-1333'', More Than a Fix in Works

AMD found itself in a tricky situation with its launch schedule for the DDR3-supportive AM3 socket platform: the revision guide document for 10h family of processors disclosed a rather serious erratum with the processors, with the company not recommending the use of more than two DDR3-1333 memory modules in all, or more than one module per memory channel, stating unreliable operation as consequence (more here). We rose doubts on a certain statement in the document circling around what we saw as a potential solution to the problem, and hence sought the company's comment.

AMD replied to us with a set of things, not necessarily interrelated, though in some way connected to the erratum, and the company's approach towards the DDR3 memory standard:
  • AMD confirms the issue as stated by the revision guide document, and is indeed working on a solution. The issue does not affect, in any way, using one DDR3-1333 memory module per channel
  • The company has already specified the safest workaround for the issue: specifying the memory modules to run at the speeds of PC3-8500, 533 MHz (DDR3-1066). One can refer to hardware literature on how to do that
  • AMD has given guidance to motherboard vendors on this issue. Some motherboards could rectify the issue by themselves, by operating the memory modules at PC3-8500 specifications. The incentive of tightening DRAM timings stays
  • Good news for overclockers: You will be able to override the memory frequency control by motherboards, and can attempt to tweak frequencies at will. AMD product warranties, as always, don't cover damages caused by overclocking, even when overclocking is enabled via AMD Software
  • Better news: Indications are that AMD is doing a lot more than creating a workaround for this issue, it may be devising a new feature altogether that, in the end, upholds the company's initiative to be enthusiast-friendly. Furthermore, AMD iterates that it is its intention to provide higher DDR3 frequencies as DDR3 becomes more prevalent
Add your own comment

19 Comments on AMD Clarifies Erratum 379 ''DDR3-1333'', More Than a Fix in Works

#1
TheMailMan78
Big Member
This is great news! I wonder what they have in store for us. I guess Dom, Binge and I wont be starting a riot after all :laugh:
Posted on Reply
#2
ShadowFold
Whoa it's like I saw into the future yesterday and it's manifesting into a news post! I hate to say it but I told you guys so :p I knew AMD was gonna get back into the game. It's awesome to see the competition right now. AMD pretty much owns the 200$ and lower market while Intel is dominating the high end. I don't see AMD getting a i7 killer out anytime soon but hey we still get fast as hell processors on the cheap! Who can complain about that?
Posted on Reply
#3
IcrushitI
increase ram volts

I don't have the equipment, but it was stated in previous articles that increasing ram voltage might be a temporary fix until amd has a permanent one. Has anyone tried it. I'm thaaaaaat close to moving up. I still hold a strong grudge at Intel for the years that they ripped me off. I look at Intel with the same disgust as Sony. I'm still boycotting both. I'm so pig headed.:)
Posted on Reply
#4
swaaye
Uhhh well the news is still that the CPUs don't work with more than a single DDR3-1333 DIMM. They just detailed the problem. They aren't going to be putting even faster DIMMs on these CPUs, I imagine. They'll probably produce a better stepping of the CPUs in a few months.

Hell, might as well stick with DDR2. I suppose the AM2+ socket has no future now though. But DDR3 gains you just about nothing with these CPUs.

It's too bad that it's apparently so hard to build support for DDR2 and 3 together in CPUs. If Intel had an i7 with DDR2 capabilities, that would be very sweet. Load one of those up with cheap DDR2 and away you go.
Posted on Reply
#5
LittleLizard
my doubt is if it affect using ddr 1600, as i read somewhere (probably here) that the am3 cpu have the multiplier to run ddr3 1600, so u just run 4 gb of ddr3 1600 and dont fuck up with slow ddr3 1333
Posted on Reply
#6
bigtye
by: LittleLizard
my doubt is if it affect using ddr 1600, as i read somewhere (probably here) that the am3 cpu have the multiplier to run ddr3 1600, so u just run 4 gb of ddr3 1600 and dont fuck up with slow ddr3 1333
Yup. agreed. This review from xbit has them running am3 cpu on 1067 and 1600.

http://www.xbitlabs.com/articles/cpu/display/phenom-ii-x4-810_6.html#sect0

The biggest issue they had was finding an am3 mobo, the 1600 ram ran fine.

Tye
Posted on Reply
#7
cdawall
where the hell are my stars
by: swaaye
Uhhh well the news is still that the CPUs don't work with more than a single DDR3-1333 DIMM. They just detailed the problem. They aren't going to be putting even faster DIMMs on these CPUs, I imagine. They'll probably produce a better stepping of the CPUs in a few months.

Hell, might as well stick with DDR2. I suppose the AM2+ socket has no future now though. But DDR3 gains you just about nothing with these CPUs.

It's too bad that it's apparently so hard to build support for DDR2 and 3 together in CPUs. If Intel had an i7 with DDR2 capabilities, that would be very sweet. Load one of those up with cheap DDR2 and away you go.
actually the chips will run DDR3 1600 just fine and i have already seen them benchmarked on XS as such. this still has very little effect on 90% of users. who uses more than 2x2GB?
Posted on Reply
#8
pentastar111
This should probably go on the post from yesterday, BUT....It often makes me wonder why they don't test this stuff out BEFORE announcing a launch?? :wtf:I mean, what did they do when testing this processor out? Assume that if it worked with 1 ram slot it would work with all of them? Everyone is in such a hurry to release this stuff...Intel does the same crap...like the ram voltage fiasco a little while ago...Video games do it ALL of the time now, constant patches! ALL of these little annoyances can be avoided all-together with ALL companies slowing it down a bit and doing a little more "in-depth" testing of these products...Then there'd be no reason for "work arounds" and constant patching....sheesh!:shadedshu It is almost as if when testing they say "Cool, it post's, get it on the market!".
Posted on Reply
#9
KBD
by: cdawall
this still has very little effect on 90% of users. who uses more than 2x2GB?
exactly!
Posted on Reply
#10
PCpraiser100
Wow, another AMD hype just like when the Phenoms were revised:laugh:
Posted on Reply
#11
hat
Maximum Overclocker
by: cdawall
who uses more than 2x2GB?
that's bs. just because 4gb is great by today's standards doesn't mean that it's ok for amd to have stability problems.
Posted on Reply
#12
eidairaman1
at least they acknowledged the problems unlike most companies.
Posted on Reply
#13
cdawall
where the hell are my stars
by: hat
that's bs. just because 4gb is great by today's standards doesn't mean that it's ok for amd to have stability problems.
go look up issues with i7 errata and come back.

http://download.intel.com/design/processor/specupdt/320836.pdf


looks like AMD is not the only one with errata's intel just kept there 21 pages of erratum under wraps instead of putting them public like AMD
Posted on Reply
#14
ShadowFold
Wow 21 pages of errata.. How embarrassing.
Posted on Reply
#15
cdawall
where the hell are my stars
by: ShadowFold
Wow 21 pages of errata.. How embarrassing.
but remember those dont count only the ones AMD makes public do.


face it people every proccessor has erratum the fact that AMD is public about it and is trying to fix it is a bloody good thing. stop going oh well it wont work with a full 4x2GB load just like P35 chipset (funny no?) doesn't insut the fact that these are good cpu's and there are currently a huge 4 models out that this even impacts the 710, 720BE, 810 and 910
Posted on Reply
#16
hat
Maximum Overclocker
by: cdawall
go look up issues with i7 errata and come back.

http://download.intel.com/design/processor/specupdt/320836.pdf


looks like AMD is not the only one with errata's intel just kept there 21 pages of erratum under wraps instead of putting them public like AMD
I never said anything about i7 errata. I said that it was wrong for you to be defending AMD, as if it was okay for them to have memory errata and nobody should care about it because nobody uses more than 2 sticks of ram anyway. I myself use 4 sticks of ram at the moment because I happened to get a very good deal on this memory kit ($90 and free shipping). Yes 8GB is overkill, but what if somebody buys a 4x1GB set of memory? Sure for some reason we consider 2x2GB sticks better than 4x1GB sticks, but it really doesn't make a considerable difference in the end.

If I make a car that for some reason doesn't work with Goodyear tires, then I am taking flak for it, but I guess I should just say that it's okay and not worry about it because there are so many other tires than only goodyear tires?

BTW I am in no way defending Intel, I myself don't like the i7 platform at all. I think it's far too expensive, and I don't like the way that if you have too much difference between your vcore and vdimm you risk frying something.
Posted on Reply
#17
roofsniper
the fact is that most people don't use more than 2 sticks of ram and even if they do just remove them or slow them down right now. a fix is on the way so hang tight and once it comes then problem solved. its really not that big of a deal im interested in what is all coming tho if its "more than a fix".
Posted on Reply
#18
Steevo
We have seen proof of a new AMD chip that runs 3.7Ghz at 1.25vcore
Posted on Reply
#19
Super XP
This is the same problem with DDR2-1066. AMD recommends one module per channel when running DDR2-1066 or faster. You can install 2 modules per channel running DDR2-800 no problem.

AMD already had this limitation, and they should have resolved this issue ASAP before the AM3 Phenom II launch. I am still waiting for that DDR2-1066 fix? It's been 2 years now?
Posted on Reply
Add your own comment