• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Intel Core Ultra 9 285k 'Arrow Lake' Discussion/News ("15th gen") on LGA-1851

Man of Honour
Joined
22 Jun 2006
Posts
12,475
Is there any chance these CPUs have the same issue as the 13th and 14th gen that are appearing or are they an entirely different architecture ?
Afaik, they're a different architecture and a new manufacturing process, but since we don't know why the 13th-14th gen CPUs are problematic, there's no way of knowing until months after release. The 13900K issues took quite awhile to be widely reported.
 
Soldato
Joined
31 Dec 2010
Posts
2,521
Location
Sussex
Afaik, they're a different architecture and a new manufacturing process, but since we don't know why the 13th-14th gen CPUs are problematic, there's no way of knowing until months after release. The 13900K issues took quite awhile to be widely reported.
Yes, optimistically the only bearing the 13/14th problems have on Arrow Lake are:
  1. Validation gets stepped up before the release
  2. Final clocks and power usage is far saner than the emergency editions - and hopefully this time if the Marketeers find out that they won't be at the top of review bars, they won't do something crazy like 5% extra performance for 50% more power.
Although for point #1, wasn't there a story that Intel got rid of their main server validation team (by forcing them to move) a few years ago, so at least then validation was seen as a low priority.
 
Caporegime
Joined
17 Mar 2012
Posts
48,495
Location
ARC-L1, Stanton System
Is there any chance these CPUs have the same issue as the 13th and 14th gen that are appearing or are they an entirely different architecture ?

You might know after 3 months of running it, or 6, or 9, or 12.......

When and if you do get it the first thing that i would do is loop it in Cinebench R24 over night, but you don't have to go that extreme, if it crashes in under 2 hours send it back.
 
Last edited:
Associate
Joined
31 Dec 2011
Posts
848
You might know after 3 months of running it, or 6, or 9, or 12.......

When and if you do get it the first thing that i would do is loop it in Cinebench R24 over night, but you don't have to go that extreme, if it crashes in under 2 hours send it back.
Yes, just interested as to whether this is going to impact the roll out of their next CPU. If there is an inherent defect it would be mad of them to add to their woes. I wasn't considering intel this time around as have been very happy with my 5900x and will be jumping on AM5 next year.
 
Last edited:
Caporegime
Joined
17 Mar 2012
Posts
48,495
Location
ARC-L1, Stanton System
Yes, just interested as to whether this is going to impact the roll out of their next CPU. If there is an inherent defect it would be mad of them to add to their woes. I wasn't considering intel this time around as have been very happy with my 5900x and still be jumping on AM5 next year.

That is a very interesting question.

I don't know what's worse, did they know these CPU's were faulty and they put them out anyway, or they didn't know.

This is not the first time Intel are out of their comfort zone in being forced to keep up with a competitor and i've been around long enough to have read the story before.

But i don't know what they would do, so..... *Shrug*
 
Last edited:
Caporegime
Joined
17 Mar 2012
Posts
48,495
Location
ARC-L1, Stanton System
Some people hold Intel in very high regard, i don't, they achieved one thing and one thing only, and that was only because IBM didn't want to give their money to Exxon being the parent company of Zilog and threatening IBM, Zilog made a better chip for the job than Intel.

Every other thing they tried has been a catastrophic failure, as i said they never made the best stuff at any point, Intel's success is directly attributed to IBM and IBM chose Intel because it was at war with the parent company of the better chip.

When Intel wanted to prove it wasn't a fluke they ended up licencing AMD's stuff and still do to this day.


 
Caporegime
Joined
17 Mar 2012
Posts
48,495
Location
ARC-L1, Stanton System
Last edited:
Soldato
OP
Joined
31 Oct 2002
Posts
9,952

5.7Ghz one core boost and 5.4Ghz all core boost on a new process (Intel 20A?) is much more conservative compared to the 6.2Ghz nonsense of 14900KS's. Far more likely to be a cooler running and stable chip - though I wouldn't put it past Intel to make a mistake, based on recent failures :D
 
Man of Honour
Joined
22 Jun 2006
Posts
12,475
Glad i'm not a retailer. I wouldn't want the hassle
From what I'm aware, AMD vastly outsell Intel in the DIY market at the moment anyway, so I don't know why they don't pull them from sale? I certainly would.


5.7Ghz one core boost and 5.4Ghz all core boost on a new process (Intel 20A?) is much more conservative compared to the 6.2Ghz nonsense of 14900KS's. Far more likely to be a cooler running and stable chip - though I wouldn't put it past Intel to make a mistake, based on recent failures :D
Didn't Rocket lake generally have lower boost/base clocks than 10th gen? It makes me wonder about what buildzoid said in how Intel had lots of knowledge of how to manage their 14nm process and maybe with their newer process they've just screwed it up.
 
Soldato
OP
Joined
31 Oct 2002
Posts
9,952
From what I'm aware, AMD vastly outsell Intel in the DIY market at the moment anyway, so I don't know why they don't pull them from sale? I certainly would.


Didn't Rocket lake generally have lower boost/base clocks than 10th gen? It makes me wonder about what buildzoid said in how Intel had lots of knowledge of how to manage their 14nm process and maybe with their newer process they've just screwed it up.
Rocket Lake increased clocks significantly from Cometlake (10th gen).
 
Soldato
Joined
22 May 2010
Posts
12,337
Location
Minibotpc
Man of Honour
Joined
13 Oct 2006
Posts
91,893
Hello?




Didn't see this coming!

Been talked about for awhile but not much enthusiasm as it seems mostly a sidegrade at best.
 
Man of Honour
Joined
13 Oct 2006
Posts
91,893
12 p core only would be interesting though,

Certainly something I'll keep an eye on, E cores are alright but not a perfect solution. For many applications I'd far rather have 12 proper cores. But I suspect it will probably just be very similar to the 14900KS for performance and power efficiency. (Some specific multi-thread workloads aside).

EDIT: What I find interesting on many of the forum threads on these Bartlett-S CPUs is people instead posting yearnings for the days of the X79 and X99 chipsets hah - kind of the same myself.
 
Last edited:
Soldato
Joined
22 May 2010
Posts
12,337
Location
Minibotpc
Certainly something I'll keep an eye on, E cores are alright but not a perfect solution. For many applications I'd far rather have 12 proper cores. But I suspect it will probably just be very similar to the 14900KS for performance and power efficiency.

EDIT: What I find interesting on many of the forum threads on these Bartlett-S CPUs is people instead posting yearnings for the days of the X79 and X99 chipsets hah - kind of the same myself.

Could never afford X79 or X99, was way out of my pricing range back then.

It makes it interesting though simply because it means i have another upgrade potential on my LGA1700 platform. Like yourself, i would much rather have a 12p core configuration and maybe more cache than e cores. Both for Gaming and editing/media work it's a great recipe!
 
Soldato
Joined
22 May 2010
Posts
12,337
Location
Minibotpc
Also interesting to see if this does come to fruition, how they will deal with the failure issues on 13th and 14th gen. From what i've read it's mostly to do with the Ring that's deteriorating right?
 
Back
Top Bottom