Intel Working on Fixing "Arrow Lake" Gaming Performance with Upcoming Patches

[AleksandarK]

In an exclusive interview with Hot Hardware, Intel acknowledged that its recently launched Core Ultra 200 desktop processors, codenamed "Arrow Lake," have significant performance issues. However, Intel announced that a set of fixes are being developed. As our review confirmed, the launch of these new processors fell short of both consumer expectations and Intel's own projections, particularly in gaming performance, despite showing promise in productivity, content creation, and some AI workloads. In a discussion during a recent livestream, Intel's Robert Hallock, VP and general manager of client AI and technical marketing, addressed these concerns head-on, describing the Arrow Lake launch as "disastrous" and attributing the underwhelming performance to inadequately optimized systems.

I can't go into all the details yet, but we identified a series of multifactor issues at the OS level, at the BIOS level, and I will say that the performance we saw in reviews is not what we expected and not what we intended. The launch just didn't go as planned. That has been a humbling lesson for all of us, inspiring a fairly large response internally to get to the bottom of what happened and to fix it.

Additionally, Hallock indicated that users can expect these updates to begin rolling out by the end of the month or shortly after that. The tech community awaits independent verification of these performance improvements, which could restore confidence in Intel's Arrow Lake platform and potentially reshape the current CPU performance hierarchy. Given that the promise is a "significant" performance uplift, we expect to see some interesting numbers as Intel's cores are performant from the microarchitectural standpoint. The mix of Windows and BIOS updates will be interesting to measure in the coming weeks. Here is the link to the video interview of the Hot Hardware crew and Robert Hallock.

View at TechPowerUp Main Site | Source

 

[Assimilator]

So now this beta product will maybe become one actually worthy of spending money one.

I will say that the performance we saw in reviews is not what we expected

Admitting that you had no idea how your own product was going to perform says things about your competence in your role, Rob - none of them good.

 

[john_]

Hallock leaving AMD to go to Intel, only to have to find excuses for Intel's incompetence.
Hope he enjoys his paycheck. By the way, is that paycheck secured?

 

[JustBenching]

So now this beta product will maybe become one actually worthy of spending money one.



Admitting that you had no idea how your own product was going to perform says things about your competence in your role, Rob - none of them good.

No fix is going to make it faster than the 14900k in games. The IMC is on a different plane of existence, it is what it is. The patch will probably address the performance on a couple of games that seem to be lower than normal (eg. cyberpunk).

Other than that I consider it a great product. It offers (already) the same gaming performance as the 9950x, at much lower power draw.

 

[Darmok N Jalad]

This was actually my biggest concern with Arrow Lake. Not the mediocre performance, but the fact that it’s a gen-1 desktop tile design. It was destined for issues, not just in terms of performance, but I wouldn’t be surprised if there are stability issues as well. Thread Director and Windows Scheduler also probably still need work. It all results in the buyer having to use with the hope that things get better. It doesn’t help that Tile might be a one-off design. Might be destined for the slag heap.

 

[ir_cow]

This will be a BIOS patch most likely to fix some internal tables that we can't see normally.

 

[Lew Zealand]

No fix is going to make it faster than the 14900k in games. The IMC is on a different plane of existence, it is what it is. The patch will probably address the performance on a couple of games that seem to be lower than normal (eg. cyberpunk).

Other than that I consider it a great product. It offers (already) the same gaming performance as the 9950x, at much lower power draw.

Most people don't want to pay more for identical performance, with a 3W power difference being negligible, not "much lower power". But in all-core, the 9950x uses 15W less power (also not a huge deal) so I don't see the reasoning?

 

[Dirt Chip]

Just add some of that “digression sauce” to juice up the pref, like you always do.
As an early 13900k adopter I know this much.

 

[sephiroth117]

Don't purchase based on a promise but actual benchmarks and stability feedbacks.

Given the price of their CPU, I really don't understand someone who wouldn't pick a 7800X3D or a 9800X3D for a gaming rig

 

[Simon.J]

I disagree with all that nonsense, their first mistake was implementation hybrid processors and the use of two different architectures, the second mistake was removing support for the AVX512 which hybrid architectures does not allow such feature because of different ISA across cores, of course, Intel lately renamed AVX512 to AVX10.2 which is already added in GCC compiler.
One more thing to add on list of shame is Intel’s mismanagement and lack of vision, I say that because they recently announced another revision of upcoming x86s which is a good thing but who knows when will it see the light, now they must move on and through away legacy which keeps them on edge.

 

[tommesfps]

As long as the games are distributed to the filler tile "cores", which are not overclockable or have L3 cache, the performance won't improve much. But the temps are fine now...

 

[Daven]

Deja vu!

 

[Fouquin]

Admitting that you had no idea how your own product was going to perform says things about your competence in your role, Rob - none of them good.

He didn't design it, he didn't build it, and he didn't do QC for it. His job is to sell a product and say whatever he can to achieve that goal. You're shitting on a man for doing his job as expected, to the letter. Be better.

 

[RayneYoruka]

Deja vu!

 

[mechtech]

By uninstalling bios security updates?

 

[windwhirl]

By uninstalling bios security updates?

This architecture just launched so I doubt there are security mitigations to consider.

 

[efikkan]

This sounds like a bunch of nonsense to me.
Let's stop pretending anything caught Intel by surprise, they have tested and validated qualification samples for months prior to release, they knew exactly the performance characteristics of the hardware on different OS'.

An updated BIOS shouldn't matter for application performance, not unless it somehow incorrectly detects or configures hardware (yes, I'm talking about the actual UEFI BIOS here, not the GUI. I find it doubtful that the EFI kernel is somehow bloated vs. Raptor Lake, not unless they've added some speedup-loops )
Tweaks to Windows' scheduler might help some workloads, and lessen the performance disadvantage compared to Linux, but is not going to fundamentally change the performance characteristics of hardware, especially for gaming.
So the only thing left would be to tweak the CPU firmware, but tightening timings here would require massive validation and might even have unintended consequences.

Software will never make up for hardware design, this only sounds like a nonsensical response to the massively largely (and undeservedly) negative response to Arrow Lake; e.g. "our new patch gains up to xx% performance", when in reality there is no significant difference.
Its application performance is great, and gaming performance on par with Raptor Lake. (And I'm of course ignoring tests at 720p/1080p with RTX 4090, as testing unrealistic workloads are pointless.)
Let's be honest about what it really is; a small step forward. If it's not big enough of an upgrade, skip it and wait for the next one. No amount of software gimmicks is going to change that.

 

[Chaitanya]

So they will get rid of rails and push the CPUs to limit throwing "efficiency" out of window, on what already is a power hungry CPU falling short of competition. A polished turd is still a turd.

 

[TheinsanegamerN]

I disagree with all that nonsense, their first mistake was implementation hybrid processors and the use of two different architectures, the second mistake was removing support for the AVX512 which hybrid architectures does not allow such feature because of different ISA across cores, of course, Intel lately renamed AVX512 to AVX10.2 which is already added in GCC compiler.
One more thing to add on list of shame is Intel’s mismanagement and lack of vision, I say that because they recently announced another revision of upcoming x86s which is a good thing but who knows when will it see the light, now they must move on and through away legacy which keeps them on edge.

People act like x86 is some kind of amp slurping monolith. It isnt. Lunar lake showed the whole "x86 cant be efficient" line was pure BS.

OTOH, removing legacy support IS one thing that will kill off X86, without legacy x86 has little future.

He didn't design it, he didn't build it, and he didn't do QC for it. His job is to sell a product and say whatever he can to achieve that goal. You're shitting on a man for doing his job as expected, to the letter.

He's in marketing, one step removed from absolute scum like used car salesman. No respect is deserved for anyone in that particular profession.

Be better.

No.

 

[Darmok N Jalad]

This sounds like a bunch of nonsense to me.
Let's stop pretending anything caught Intel by surprise, they have tested and validated qualification samples for months prior to release, they knew exactly the performance characteristics of the hardware on different OS'.

An updated BIOS shouldn't matter for application performance, not unless it somehow incorrectly detects or configures hardware (yes, I'm talking about the actual UEFI BIOS here, not the GUI. I find it doubtful that the EFI kernel is somehow bloated vs. Raptor Lake, not unless they've added some speedup-loops )
Tweaks to Windows' scheduler might help some workloads, and lessen the performance disadvantage compared to Linux, but is not going to fundamentally change the performance characteristics of hardware, especially for gaming.
So the only thing left would be to tweak the CPU firmware, but tightening timings here would require massive validation and might even have unintended consequences.

Software will never make up for hardware design, this only sounds like a nonsensical response to the massively largely (and undeservedly) negative response to Arrow Lake; e.g. "our new patch gains up to xx% performance", when in reality there is no significant difference.
Its application performance is great, and gaming performance on par with Raptor Lake. (And I'm of course ignoring tests at 720p/1080p with RTX 4090, as testing unrealistic workloads are pointless.)
Let's be honest about what it really is; a small step forward. If it's not big enough of an upgrade, skip it and wait for the next one. No amount of software gimmicks is going to change that.

Nah, this is a new architecture. They eventually have to release the product, knowing full well that they could get more out of it. If they wait too long, they miss their launch window and sale seasons. They very likely got Arrow Lake to a mostly stable place and put them on the market. I don’t doubt they have things to work on, and I would not be surprised if they get more performance out of the product. That said, I wouldn’t buy this product on such hopes.

 

[docnorth]

Most people don't want to pay more for identical performance, with a 3W power difference being negligible, not "much lower power". But in all-core, the 9950x uses 15W less power (also not a huge deal) so I don't see the reasoning?

Referring to TPU’s review, 285K consumes on average 3w less during applications and 10w less during gaming, so basically negligible difference. Idling otoh shows a 22w difference and this trend (not strict tests, just comparing my 13700 with a friend’s new 7800x3d) seems to continue during low load. But for Intel‘s marketing and image the big problem are the x3d chips and it will remain unsolved even if they manage to reach 14900K for gaming, which seems unlikely.

 

[Ravenmaster]

Well first they need to fix their prices by lowering them because nobody is going to buy their Core Ultra 200 series chips while AMD's 9800X3D exists.

 

[HisDivineOrder]

I imagine a world where Intel only released processors with the Pcores. They had real gains. Ecores is a boondoggle that's led them farther and farther away from being competitive. They could still use them in some kind of CPU with tons of them built for servers and possibly consoles.

But they should not be expecting enthusiasts to be trying to run games on Ecores instead of just every single game thread being on a Pcore. Eight threads is not enough for modern games ported from CPU's with potentially 16 threads, which means games will inevitably drop back to Ecores and suddenly you'll have screwy 1% lows.

Which is exactly what is happening. You see it on their chips and you see it on the Steam Deck with its 8 threads with four off hyperthreading. That's why inevitably Valve will swap to a 6 core/12 thread or 8 core/16 thread CPU for the next one. They'll see the writing on the wall. Why didn't Intel?

Whoever at Intel thought a high performance enthusiast chip could run off only 8 pcore threads should have been fired yesterday.

 

[Metroid]

They want trolls to keep buying their crap then they will say, hey we never said there was something to be fixed ehhe

 

[phanbuey]

This was actually my biggest concern with Arrow Lake. Not the mediocre performance, but the fact that it’s a gen-1 desktop tile design. It was destined for issues, not just in terms of performance, but I wouldn’t be surprised if there are stability issues as well. Thread Director and Windows Scheduler also probably still need work. It all results in the buyer having to use with the hope that things get better. It doesn’t help that Tile might be a one-off design. Might be destined for the slag heap.

Not to mention there was huge regression in the last 6 months for all hybrid designs. The 14900K went from just behind the 7800x3d to pretty far down on the charts in newer reviews.

Something is going on with microsoft/intel and the thread scheduling.