News Intel's Arrow Lake performance fix is now available — another update coming next month

This is good. I'm looking forward to the test results.
For starters not having a properly functioning Windows power plan on a hybrid CPU could make a big difference. What a mess of a launch if the performance goes up as much as claimed.

It might have had something to do with Gelsinger's departure.
 
This is good. I'm looking forward to the test results.
For starters not having a properly functioning Windows power plan on a hybrid CPU could make a big difference. What a mess of a launch if the performance goes up as much as claimed.

It might have had something to do with Gelsinger's departure.
I agree that it was a mess of a launch but I can't imagine Gelsinger leaving had too much of an impact. Can't imagine what a CEO is needed for a launch that's been planned for months. To me its just they rushed things out a little premature, they really wanted it out by a specific date regardless of kinks.
 
  • Like
Reactions: artk2219
The most recent BIOS updates from MSI have been available in beta form since the end of Oct. Will have to check on the MS KB as I purposely installed Win 11 23H2 when setting up my system. I've done Windows updates and a BIOS update since my first 3DMark test and the CPU score went up a fair bit between the two despite identical clocks so something in there did help out.
threw in a run from the old system for fun
dDciF8L.jpeg

I don't really expect any of these things to make a real overall difference, but if they can knock down the outliers that would be good.

While they're at it Intel should really do something about the absurdity of installing the management engine. Unless it has changed recently the only simple way to install it is through Windows (when I had to do it on my server box it was faster to just put a drive with Windows on it in and boot that to install).
 
The Windows build was mentioned multiple times by Intel and this article, but it would have been clearer if they added the build translation of Windows 11 24H2 as some might have updated to December's update but I assume would not have Intel's updates if still on 23H2, which many folks still would be.

Intel wants their own brand of Fine Wine like AMD these days, a?
 
Fear that despite this, and whatever gains or stability improvements result, the damage is done to their reputation, especially after the raptorlake cpu failures. Its also highly unlikely that arrow lake is going to notch wins against the beastly 9800X3d.

"Intel's Application Performance Optimizer (APO) boosts game performance in several game titles. This software utility is now automatically installed by default in Windows"
-Don't have to install an app to make the CPU work properly on AMD
 
Last edited:
  • Like
Reactions: bit_user
Fear that despite this, and whatever gains or stability improvements result, the damage is done to their reputation, especially after the raptorlake cpu failures. Its also highly unlikely that arrow lake is going to notch wins against the beastly 9800X3d.

"Intel's Application Performance Optimizer (APO) boosts game performance in several game titles. This software utility is now automatically installed by default in Windows"
-Don't have to install an app to make the CPU work properly on AMD
IMO arrow lake does not need to beat the 9800X3D, it just needs to beat the 13000 and 14000 series, and the 7000 and 9000 series non-X3D CPUs.
 
What a mess of a launch if the performance goes up as much as claimed.
Yes

It might have had something to do with Gelsinger's departure.
I doubt it made a big impact in the board's overall sentiment. Didn't Intel still manage to sell out of most Arrow Lake models, in spite of their poor launch performance numbers?

It is kinda weird to see Qualcomm, AMD, and Intel all fumbling their launches this badly!
 
Last edited:
  • Like
Reactions: thestryker
Yes


Uh, no. That happened after Arrow Lake launched!

It's plausible the 15% layoffs could've had an effect. However, even that seems like it came too late to disrupt their launch plans for Arrow Lake.

I fear the rot goes even deeper, but the scale of this miss should have focused minds on the problem. I doubt Intel's next launch will be handled this badly.

It is kinda weird to see Qualcomm, AMD, and Intel all fumbling this badly!
Is it possible the original poster meant that the launch issues were partially the reason Gelsinger was ousted?
 
  • Like
Reactions: bit_user
I agree that it was a mess of a launch but I can't imagine Gelsinger leaving had too much of an impact. Can't imagine what a CEO is needed for a launch that's been planned for months.
I'm pretty sure he meant it the other way around. Gelsinger left about a month after the launch, so there's no way it caused Arrow Lake's bad launch.

Maybe the 15% layoffs had some impact, but I think most of the planned prep work on Arrow Lake's launch would've been completed, prior to when that hit.

To me its just they rushed things out a little premature, they really wanted it out by a specific date regardless of kinks.
That seems the norm, for this industry. If it had indeed been merely "kinks", I doubt anyone would've really minded. I think these problems rise well above the level of mere "kinks".

Is it possible the original poster meant that the launch issues were partially the reason Gelsinger was ousted?
I had already revised my post. You just beat me to it!
: D
 
  • Like
Reactions: helper800
So now even CPUs need a zero-day patch...
That's not unusual. If you read lots of launch-day CPU reviews, they often note issues they encountered. Many times, this is due to using pre-release BIOS and sometimes Intel/AMD even manage to turn around a fix in time for the review's completion.

These fixes took like 2 months, though. Stuff definitely wasn't done in time for launch that should've been. I have to wonder whether that was due to cancelling the 20A node, which might've meant they had far less time than usual with silicon in house, prior to launch.

It's probably a good thing the socket wasn't new and that they reused Meteor Lake's I/O tile. There are some Meteor Lake-S engineering samples floating around, so I'm sure those were used to develop & debug basic platform support, before Arrow Lake silicon ever arrived.
 
I would buy 265K over 9700X/9800X3D if on average 265K was faster than 9700X in gaming and on average faster than 9900X in productivity. I want more than 8 cores for productivity and as long as gaming hasn't regressed from Raptor Lake, I don't bleeding edge gaming performance.

Still I'm waiting to see what 9900X3D offers. If they resurrect this no-mans land cpu so it no longer sucks at productivty like 7900X3D and was noticeably weaker than 7800X3D for gaming, it might tempt me. 9800X3D is a surprise packet with productivity performance as good or better than 9700X in most cases. I want to see that reproduced with 9900X3D as I have no interest in 9950X/X3D.
 
  • Like
Reactions: bit_user
Fear that despite this, and whatever gains or stability improvements result, the damage is done to their reputation, especially after the raptorlake cpu failures. Its also highly unlikely that arrow lake is going to notch wins against the beastly 9800X3d.

This sadly is likely the case, a lot of loyal Intel customers who used to buy Intel anyway due to the traditional superior stability and reliability was lost in the past season as the RPL disaster is admitted and bombed around. It worsens when the Ultra line with a branding change don't really out perform the RPL much in production and with gaming regression. It’s evident how retail AMD vs Intel sales are recently
 
I agree that it was a mess of a launch but I can't imagine Gelsinger leaving had too much of an impact. Can't imagine what a CEO is needed for a launch that's been planned for months. To me its just they rushed things out a little premature, they really wanted it out by a specific date regardless of kinks.

That's not unusual. If you read lots of launch-day CPU reviews, they often note issues they encountered. Many times, this is due to using pre-release BIOS and sometimes Intel/AMD even manage to turn around a fix in time for the review's completion.

These fixes took like 2 months, though. Stuff definitely wasn't done in time for launch that should've been. I have to wonder whether that was due to cancelling the 20A node, which might've meant they had far less time than usual with silicon in house, prior to launch.

It's probably a good thing the socket wasn't new and that they reused Meteor Lake's I/O tile. There are some Meteor Lake-S engineering samples floating around, so I'm sure those were used to develop & debug basic platform support, before Arrow Lake silicon ever arrived.
I think the launch was just the icing on the cake for the series of mishaps that lead to Arrow Lake's unexpectedly poor performance. The 20A not working, having TSMC do the core tile, using more tiles than reasonably necessary contributing to the memory controller being off the compute die. Then not even having Windows being able to interact with the chip properly.

Just a long series of failures in execution in the production of a core product. If Intel would have used a refined Intel7, kept the clocks a little lower for thermal headroom, the IPC and memory controller improvements would have made a far better CPU for a lot less money than what was made. Those cores are a real step up, they just are shackled to a relative lemon of a platform. IDK, maybe some of the piecing together of chiplets is to showcase the foundry and packaging possibilities for potential IDM 2.0 customers.
 
If Intel would have used a refined Intel7,
How much more refinement do you think is even still possible? Raptor Lake is already using like their 5th gen 10nm DUV node! Canon Lake used the first gen, Ice Lake was second gen (10 nm+), Tiger Lake was 3rd gen (SuperFin), Alder Lake is 4th gen (Enhanced SuperFin), and Raptor Lake is 5th gen (Enhanced SuperFin+).

kept the clocks a little lower for thermal headroom, the IPC and memory controller improvements would have made a far better CPU for a lot less money than what was made.
Using Intel 7 wouldn't have enabled them to provide the substantial IPC increases of Skymont and Lion Cove. The resulting cores would've been too big and power-hungry, leading to a "Rocket Lake" situation.

Also, Intel 7 was lagging on efficiency for laptop products, which is why they felt a need for Meteor Lake to swerve so sharply in the direction of improving efficiency.

I would agree that maybe Intel should've tried harder to use Intel 3 for Lunar Lake and Arrow Lake. I honestly wonder what the performance and efficiency differences would've been, compared to TSMC N3B.
 
  • Like
Reactions: thestryker
I would agree that maybe Intel should've tried harder to use Intel 3 for Lunar Lake and Arrow Lake. I honestly wonder what the performance and efficiency differences would've been, compared to TSMC N3B.
If the reason for the 20A cancelation is what they stated then there's no way they would have had the capacity for ARL on Intel 3 (that if is obviously doing the heavy lifting here). I'm really curious what they're going to use Intel 3 for in house and hope we get to see some client parts on it just so we can get an idea. I could see a world where Intel does lower SKUs on Intel 3 and uses 18A+ for the higher margin parts (I imagine Intel 3 would be pretty ideal for E-core only parts).
 
  • Like
Reactions: bit_user
If the reason for the 20A cancelation is what they stated then there's no way they would have had the capacity for ARL on Intel 3 (that if is obviously doing the heavy lifting here). I'm really curious what they're going to use Intel 3 for in house and hope we get to see some client parts on it just so we can get an idea. I could see a world where Intel does lower SKUs on Intel 3 and uses 18A+ for the higher margin parts (I imagine Intel 3 would be pretty ideal for E-core only parts).
Maybe in the future it could be doing the less node sensitive chiplets, who knows.

What I hope is they could get into their battlemage style of products. Sometimes it's not really about the Halo product, but what the C/P ratio does
 
Even if Intel fixes their CPU problems , They need to be punished for releasing beta version of their CPU ... Where were the Testers and the Engineering samples ? what are you doing intel ? did you even BOTHER to test your products before releasing them . NO THANK YOU. you need to be punished.
I hope ALL skip this generation. intel really needs a hard lesson.
 
  • Like
Reactions: endocine
Even if Intel fixes their CPU problems , They need to be punished for releasing beta version of their CPU ... Where were the Testers and the Engineering samples ? what are you doing intel ? did you even BOTHER to test your products before releasing them . NO THANK YOU. you need to be punished.
I hope ALL skip this generation. intel really needs a hard lesson.
nudge nudge...Faildozer.
 
Even if Intel fixes their CPU problems , They need to be punished for releasing beta version of their CPU ... Where were the Testers and the Engineering samples ? what are you doing intel ? did you even BOTHER to test your products before releasing them . NO THANK YOU. you need to be punished.
Well, it's your money, so do what you want with it.

I think people should mostly just make decisions in their self-interest, and if that means skipping this generation because it's not a good fit for your priorities, then so be it.

As for punishment, I think Intel does seem to have gotten the message.

I hope ALL skip this generation. intel really needs a hard lesson.
I will be skipping Arrow Lake, but only for the simple reason that my last Intel CPU purchase was Alder Lake, which is still more than good enough for me. I'd rather hold out for Nova Lake, because that should introduce APX and AVX10, which are novel features. Also, it should hopefully reunite the memory controller with the compute tile, as happened in Lunar Lake.
 
Last edited: