"Issues" with Conroe ES chips to be rectified in retail version [New stepping]

[Super Nade]

"Issues" with Conroe ES chips to be rectified in retail version [New stepping]

Folks,
I came accross this:
http://www.hkepc.com/bbs/itnews.php?tid=627088

Intel hints that Conroe is going to release at B-2 Stepping as Intel Core 2 Duo processor. As for the previous version, a problem was found to make the system full loaded. It’s only solved in the new stepping. We don’t encourage anyone to buy the engineering sample from the web. The retail version is going to release in the end of this month, and it’s much stable.

Further, some of the manufacturers noticed that the released Intel P965 chipset has got the expected performance from Fast Memory Access technology, performing the same as i975X with DDR2 800. Intel has promised to fix the problem in the next P965 C-2 Stepping, and is expected to release in late July.

With regard to the reports on the weak performance of G965, sources indicate that it is because the driver hasn’t been finished. It’s not the stage to make any conclusion yet. Intel scheduled to add DVMT 256MB support in v14.21, releasing on July 13, Pixel Shader 3.0 and T&L support in v14.24, and Hardware Geometry (Vertex Shader 1.1/2.0) support in v14.26, where v14.26 is expected to be the final version. Besides, G965 has stepped in C-1 Stepping, yet there are still some problems to be fixed. We expected G965 would be releasing at C-2 Stepping. It’s very soon to the releasing day (July 23), but it seems the progress is still far away from expected. Whether G695 could release on time is a big question!

What does he mean by having probblems when full system loaded? Is it problem when the system is loaded or is it when the system is not loaded? Any of you ES boys see any problems?

I presume that memory issues are not limited to the P version of the chipset. So end of July it is, eh?

 

[Gautam]

Well, too late for that. I don't seem to have any problem "to make the system full loaded." Still buying an ES is a dumb choice for most, I only did it because I'll be back at school without a computer shortly after release and retail Conroes won't do me any good.

 

[donv]

No "system load" troubles for me, as of yet. This is an E6300, in the process of being clocked.

It sounds rather vague, more like propaganda to stop these chips being sold before release than anything else. Intel already seemed to have forced 3dmark too take down Conroe scores from the ORB, so i guess they aren't 100% happy about them being so commonplace in the wilderness.

 

[Gautam]

No way would Intel have FM remove Conroe scores. #1 3DMark ranks across the board would be nothing but free advertising for them. Though for sure they aren't pleased with them being available in the wild.

 

[veryhumid]

is "a problem was found to make the system full loaded" even grammatically correct?

 

[DavidJa]

maybe regarding the raid slowdown problems?

 

[donv]

No way would Intel have FM remove Conroe scores. #1 3DMark ranks across the board would be nothing but free advertising for them. Though for sure they aren't pleased with them being available in the wild.

Well there was a big debate about this at XS. But the fact stands, one day, for some reason, all the conroe scores were taken off the ORB (aside from in 3dmark01, which is now 'unsupported'. And no reason was given, and to my knowledge, still hasn't been.

 

[Burdman27911]

maybe regarding the raid slowdown problems?

But wasn't that a chipset problem? IDK I'm all confused today, at least it's a friday...

 

[Gautam]

Well there was a big debate about this at XS. But the fact stands, one day, for some reason, all the conroe scores were taken off the ORB (aside from in 3dmark01, which is now 'unsupported'. And no reason was given, and to my knowledge, still hasn't been.

Trust me, I'm fully aware as I've got a pretty kickass suite of scores waiting to be uploaded...I'm simply dropping a hint as to who else's best interests dropping Conroe scores from the ORB would be in...

 

[donv]

DAM i see what you're on about now.

 

[Super Nade]

hkepc is a Chinese/Tiawanese site. Maybe we should overlook the grammatical mistakes. I'm still puzzled by the "problems", they speak of. If there were any, our ES boys/girls would have reported them.

May be BS propoganda or maybe there is something to it which is lost in translation. As long as I don't see a test or numbers, to me, this falls in the relam of vivid speculation.

 

[RangerXLT8]

Trust me, I'm fully aware as I've got a pretty kickass suite of scores waiting to be uploaded...I'm simply dropping a hint as to who else's best interests dropping Conroe scores from the ORB would be in...

Are you saying that Intel is after folks posting Conroe bEnchmarks on ORB? Or am I completely misunderstanding you?

 

[Gautam]

What I'm saying is that whoever is at the top of the ORB gets free advertising. ATM, Intel is not at the top of the ORB because Conroe scores are pulled. Someone else is. Take a wild guess at which company that would be. Now this is just total speculation but its the only conclusion my cynical mind can come to...I've been an avid benchmarker for years and have very often seen prelease scores published to the ORB, and have never seen them removed. This is the first time ever, which makes me think that there is more than simple Futuremark policy at play here (their official answer). But don't take much heed of my thoughts, they're mine only.

 

[Stratcat]

Between the translation and possible Intel FUD, IMO, it's near impossible to read much of anything firm from that article.

But I do know one thing: A new stepping from Intel generally means better performance and/or features, in some aspect of the product's abilities (i.e C0, D0, EO Prescotts).

So I, for one, believe this is a good thing.

 

[donv]

No-one's denying that it's a good thing.

It's just that i've failed to notice any thing going wrong at full load. And i've run 2xsuperpi32m while installing a game and burning a CD, so my E6300 ES has been fairly well loaded. In fact, i've failed to notice anything in particular wrong at all. That's the mystery.

 

[Stratcat]

[...]

In fact, i've failed to notice anything in particular wrong at all. That's the mystery.

That's why it's called FUD!

 

[donv]

What does FUD mean, like nonsense? Sounds like an "Inquirer-ism" to me...

 

[BigDan3131]

My buddy from Boeing was hearing these rumors too but what he heard was RAID 5 had serious issues with the current samples.

 

[Stratcat]

What does FUD mean, like nonsense? Sounds like an "Inquirer-ism" to me...

FUD = Fear, Uncertainty, Doubt.

http://en.wikipedia.org/wiki/FUD

But funny you mention the Inquirer: The inquirer recently had a blurb linking the OP's HKEPC article.

 

[donv]

I know, i've seen this all over the place.

@Bigdan;
Wouldn't RAID5 be a motherboard issue rather than CPU? Cos i heard that the US goverment declined to use Kentsfield, due to the fact the 975X Raid5 support was crap.