*Official Benchmark Thread* - Post it here or it didn't happen :D

[Mr. Fox]

23 minutes ago, Meaker said:

 

The 9950X3D with the improved cache die layering and 3d vcache on both CCDs could be the most interesting high end enthusiast chip for a while.

Yes it certainly could be if it overclocks better than a 9950X does. I'm not very interested in stock performance though, because a CPU or GPU that doesn't overclock well is very boring and of zero value to me even if it is the most powerful option available for purchase. The fact that it will overclock at all will be an improvement over previous X3D chips since they were locked down.  I'm not very optimistic but open-minded when it comes to AMD. I am not a brand-loyalist, but I do hold a grudge when I'm not happy about things turn out.

 

Let's cross our fingers and hope they do something different and better that what Brother @Raiderman described about his experience with 9950X, which sounds remarkably similar to (maybe not as bad as) the new Core Ultra rubbish. I don't buy hardware on the basis of it being better for gaming and don't care too much about that.

[Raiderman]

56 minutes ago, Mr. Fox said:

Yes it certainly could be if it overclocks better than a 9950X does. I'm not very interested in stock performance though, because a CPU or GPU that doesn't overclock well is very boring and of zero value to me even if it is the most powerful option available for purchase. The fact that it will overclock at all will be an improvement over previous X3D chips since they were locked down.  I'm not very optimistic but open-minded when it comes to AMD. I am not a brand-loyalist, but I do hold a grudge when I'm not happy about things turn out.

 

Let's cross our fingers and hope they do something different and better that what Brother @Raiderman described about his experience with 9950X, which sounds remarkably similar to (maybe not as bad as) the new Core Ultra rubbish. I don't buy hardware on the basis of it being better for gaming and don't care too much about that.

The 9950x is still a beast of a chip. I think I may be able to hit 47k in CB 23, and 246K mips in 7zip. I need to check the waterblock contact.

[electrosoft]

1 hour ago, Meaker said:

 

The 9950X3D with the improved cache die layering and 3d vcache on both CCDs could be the most interesting high end enthusiast chip for a while.

 

1 hour ago, Mr. Fox said:

Yes it certainly could be if it overclocks better than a 9950X does. I'm not very interested in stock performance though, because a CPU or GPU that doesn't overclock well is very boring and of zero value to me even if it is the most powerful option available for purchase. The fact that it will overclock at all will be an improvement over previous X3D chips since they were locked down.  I'm not very optimistic but open-minded when it comes to AMD. I am not a brand-loyalist, but I do hold a grudge when I'm not happy about things turn out.

 

Let's cross our fingers and hope they do something different and better that what Brother @Raiderman described about his experience with 9950X, which sounds remarkably similar to (maybe not as bad as) the new Core Ultra rubbish. I don't buy hardware on the basis of it being better for gaming and don't care too much about that.

 

Both the 9800X3D and 9950X3D are interesting because they're putting the 3DVcache underneath the CCDs vs on top which was a limiting factor before due to their sensitivity and location. This also hindered the CCDs from a good old proper OC.

 

Couple this with *if* both CCDs are getting the 3DVcache on the 9950X3D (still rumored last I checked) along with even better CCD to CCD sync (aka no AMD dip), the 9950X3D could be the monster to beat in every aspect.  I'm waiting to see how the CCDs are binned vs the 9950X where you have a good one and a slightly subpar one.

 

I'm really hoping it delivers especially with that "chunk" you get when the cache gets full as it has really shown up in WoW especially with skyriding vs Intel. Every location where I would get that chunk especially in raids is gone now or vastly reduced to the point of being a Blizzard issue. Spending a week or so going back and forth for hours on end playing (or the tragedy!) and logging every time it happened, it was very noticeable.

 

The kicker was playing on the non X3D CCD only for hours. Less performance definitely but that chunk was basically gone too. That was something I hadn't tested before but since I was on a 7950X3D, I had the ability to do that. That moved it from "AMD Dip issue" to "Cache saturation issue" in my book.

 

Also toss In the fact with  WoW, it is hard to differentiate between Blizzard engine induced microstutters because of high frame rates and/or high player/asset areas and Config issues.

 

 

[Mr. Fox]

3 hours ago, Raiderman said:

The 9950x is still a beast of a chip. I think I may be able to hit 47k in CB 23, and 246K mips in 7zip. I need to check the waterblock contact.

Yes, it is a strong CPU. I haven't ruled it out completely as an option to consider, I am just very apprehensive about a lot of things based on a long history of disappointment. But, for the first time ever I am disappointed with what Intel brought to the table. I don't like the design of Ryzen architecture, but it looks like the writing is on the wall now with Intel adopting tiled/chiplet crap. The choice isn't as clear as it was before Core Ultra reared its ugly head. I know you and I have discussed how a Ryzen 9 CPU might behave on a chiller since it is thermally limited boost clock. (Behavior is more like a modern GPU.) That could prove to be interesting if boost clock scales up when cool as much as it scales down when not cool.

 

I'm leaning toward just doing nothing as being the best option since I am not concerned in the least about i9 degradation. But, no need for urgency, especially with Black Friday and Christmas sales coming soon. And who knows... maybe the bottom will fall out on Zen 5 and Core Ultra platform pricing since both have had a cooler reception than AMD and Intel had hoped they would.

  

2 hours ago, electrosoft said:

Both the 9800X3D and 9950X3D are interesting because they're putting the 3DVcache underneath the CCDs vs on top which was a limiting factor before due to their sensitivity and location. This also hindered the CCDs from a good old proper OC.

 

Couple this with *if* both CCDs are getting the 3DVcache on the 9950X3D (still rumored last I checked) along with even better CCD to CCD sync (aka no AMD dip), the 9950X3D could be the monster to beat in every aspect.  I'm waiting to see how the CCDs are binned vs the 9950X where you have a good one and a slightly subpar one.

 

I'm really hoping it delivers especially with that "chunk" you get when the cache gets full as it has really shown up in WoW especially with skyriding vs Intel. Every location where I would get that chunk especially in raids is gone now or vastly reduced to the point of being a Blizzard issue. Spending a week or so going back and forth for hours on end playing (or the tragedy!) and logging every time it happened, it was very noticeable.

 

The kicker was playing on the non X3D CCD only for hours. Less performance definitely but that chunk was basically gone too. That was something I hadn't tested before but since I was on a 7950X3D, I had the ability to do that. That moved it from "AMD Dip issue" to "Cache saturation issue" in my book.

 

Also toss In the fact with  WoW, it is hard to differentiate between Blizzard engine induced microstutters because of high frame rates and/or high player/asset areas and Config issues.

It will be interesting to observe if any new drama unfolds with the Zen 5 X3D chips. For whatever silly reason, it seems like new tech has an unhealthy dependency on Micro$lop's abortion OSes being optimized for proper functionality. When you stop and think about it, that is extremely sucky and shouldn't be the case at all. Every time AMD releases something new there is something wrong with Windoze holding it back that has to be fixed, and then when Intel started the smartphone core integration it was thread scheduler this, thread scheduler that, blah, blah, blah.  None of that actually mattered using manual BIOS settings (fixed clocks and voltages) for everything... Winduhz became irrelevant. Windows 7 and 10 work marvelously with no Winduhz 11 thread scheduler nonsense with manual settings. Not sure if that applies equally to Ryzen as well.

[Mr. Fox]

Hey guys. If you haven't seen this benchmark tool, check it out. I saw it on a Buildzoid video.

 

https://github.com/clamchowder/MicrobenchmarksGui/releases/tag/v0.000006

 

It can be used to benchmark system RAM bandwidth and latency, and GPU VRAM when copying from CPU to GPU and GPU to CPU memory.

 

[electrosoft]

I went ahead and cancelled my 285k order. Sanity prevailed and I'm going to return to my "wait and see" approach to see how it pans out over the next few months. With the 9950X3D slated early next year, we should see whatever refinements / the OC community can tune up by then for comparison.

 

Plus...you know....5090 Wallet Slayer Editions will be hitting the market most likely in Q1 too. Have to have a sufficient buffer to properly be able to yay or nay that Balance Destroyer.

 

 

[tps3443]

8 hours ago, electrosoft said:

I went ahead and cancelled my 285k order. Sanity prevailed and I'm going to return to my "wait and see" approach to see how it pans out over the next few months. With the 9950X3D slated early next year, we should see whatever refinements / the OC community can tune up by then for comparison.

 

Plus...you know....5090 Wallet Slayer Editions will be hitting the market most likely in Q1 too. Have to have a sufficient buffer to properly be able to yay or nay that Balance Destroyer.

 

 

Can you imagine if Nvidia released a 5090, and they are like. “It has less cores” “It uses less power” it’s up to 5-10% faster than 4090” “And It’s more expensive” 🤣 I think you made the right move cancelling the 285K. These platforms as a whole similarly priced to high end GPU's, if we account for top notch motherboards + CPU + Ram. I think we can count on the 5090 being a beast. 

[Mr. Fox]

2 hours ago, tps3443 said:

Can you imagine if Nvidia released a 5090, and they are like. “It has less cores” “It uses less power” it’s up to 5-10% faster than 4090” “And It’s more expensive” 🤣 I think you made the right move cancelling the 285K. These platforms as a whole similarly priced to high end GPU's, if we account for top notch motherboards + CPU + Ram. I think we can count on the 5090 being a beast. 

But, will it overclock? That's the million dollar question. With GPUs we are moving in the direction of just barely already because they are so castrated by firmware. If it is a beast and doesn't overclock, then "do not pass go and do not collect $200" is how I look at it. It doesn't earn the title "beast" if it doesn't overclock no matter how powerful it is, it's just a worthless gamerboy consumer play toy to me. No overclock = "push the buttons, pull the chain, out comes chocolate choo-choo train" or, if you prefer, "milk, milk, lemonade, 'round the corner fudge is made" 🤣

[cylix]

AMD Ryzen 7 9800X3D Overclocked to 5.46 GHz, Beating Ryzen 7 7800X3D by 24%

[Mr. Fox]

7 hours ago, cylix said:

AMD Ryzen 7 9800X3D Overclocked to 5.46 GHz, Beating Ryzen 7 7800X3D by 24%

It is awesome to see that kind of performance jump. Anyone that doesn't like that is either lying or stupid.

 

Speaking of stupidity, I think one of AMD's biggest challenges is the mentality of some of their customers. Like this one... That is detrimental to all concerned, including those that don't care but are too dumb to know better.

 

[electrosoft]

3 hours ago, tps3443 said:

Can you imagine if Nvidia released a 5090, and they are like. “It has less cores” “It uses less power” it’s up to 5-10% faster than 4090” “And It’s more expensive” 🤣 I think you made the right move cancelling the 285K. These platforms as a whole similarly priced to high end GPU's, if we account for top notch motherboards + CPU + Ram. I think we can count on the 5090 being a beast. 

 

I'd be somewhat excited if they were talking about a super efficient variant for specific laptops but everything else? Nah....

 

I was hoping with 285k and it running a least more efficient along with the 7800/9800 X3D variants we would see the return of proper DTRs or at least one model but the NH55 really does seem to be the end of the road for desktop CPU support. 😞

 

 As I was adding up the pricing of the 285k, a decent motherboard and CU Dimms, I was quickly over $1300 and climbing. I just went, "whoa whoa whoa...." Toss in a 5090 and suddenly I'm looking at $3k+ and most likely 3.5k+.

 

Plus I still have my complete other AM5 system I'm holding onto besides having sold my Powercolor 7900XTX. Since my daughter abandoned it (lol), I'm going to rebrand it into an anime themed chassis and that leaves open the chance to slap in a 9800X3D or 9950X3D to play around with in the future. Right now, it is rocking a 7600x. If my daughter suddenly gets the urge to play something else, I can always swap in another GPU as needed. I just want to keep at least one AM5 platform in the house going "just in case"

 

 

Unless someone offers a decent price, I'm going to ride my 4090 up to and through the first part of getting a 5090 and eliminating bad samples/coil whine units till i get something decent. 2+ yrs of solid service from the 4090 can't be beat. Then it has to provide a real, tangible upgrade in 4k from the 4090 calculating in not much uplift from CPUs (9000 X3D testing of course needed). From there, I might slap it in my wife's system to upgrade her 3080 10gb which would be pretty monstrous. Or when all is said and done, I might ride it to the 6090! 🤣

 

 

[Papusan]

5 hours ago, electrosoft said:

Right now, it is rocking a 7600x. If my daughter suddenly gets the urge to play something else, I can always swap in another GPU as needed. I just want to keep at least one AM5 platform in the house going "just in case"

 

That one cost close to 1/3 of an Core Ultra 9 285K and perform about equal in games. On top.... Intel pay more for TSMC silicon and it is still worse than the cheapest you get from AMD. Hmmm. Nothing can save Arrow lake from an utter flop. In short... Intel have nothing for gaming outside their old soon EOL in-house made Raptor lake chips. 

 

Even these are shorter in the tooths now. No point in buying a brand new or the old Intel platform for tomorrow's gaming. So Intel offer nothing for the gamers. Only loads of power efficient Baby cores and castrated P-cores (no HT).

 

But the new tile mess from Intel isn't bad in everything for gaming...🥴 Intel Core Ultra 9 285K internal graphics can play Battlefield 2 at an average of 120+ FPS at 1080p medium and low quality.

 

Here's more from today's tech trend....

I've tested Asus' new Turbo Game Mode for AMD CPUs and my verdict is simple: Don't use it

[electrosoft]

Both tuned, gaming wise 14900k still leads overall vs 285k: https://quasarzone.com/bbs/qf_cmr/views/2566736 but it's fairly close overall. Or close enough and comes down to a game by game comparison. Like I said, 285k can only get better from here while we're seeing the 14th gen at its best.

 

I do recognize and appreciate all the other architectural gains starting with power consumption overall and lower clocks still providing competitive performance. This sets a great baseline for future enhancements from Intel. It smells of P4 vs P3 all over again where those first gen P4 chips were fighting hard to compete with the last gen but eventually stretched their legs for many reasons and at the end clearly outclassed P3.

 

As for Windows, I do not know if they will need to figure out a way to redesign the scheduler from the ground up for Windows 12 (shudder), but it clearly still leans towards homogenized monolithic designs and asking it to figure out how to prioritize continues to plague it for performance outside of full tilt multicore loads.

 

48 minutes ago, Papusan said:

 

That one cost close to 1/3 of an Core Ultra 9 285K and perform about equal in games. On top.... Intel pay more for TSMC silicon and it is still worse than the cheapest you get from AMD. Hmmm. Nothing can save Arrow lake from an utter flop. In short... Intel have nothing for gaming outside their old soon EOL in-house made Raptor lake chips. 

 

Even these are shorter in the tooths now. No point in buying a brand new or the old Intel platform for tomorrow's gaming. So Intel offer nothing for the gamers. Only loads of power efficient Baby cores and castrated P-cores (no HT).

 

But the new tile mess from Intel isn't bad in everything for gaming... Intel Core Ultra 9 285K internal graphics can play Battlefield 2 at an average of 120+ FPS at 1080p medium and low quality.

 

Here's more from today's tech trend....

I've tested Asus' new Turbo Game Mode for AMD CPUs and my verdict is simple: Don't use it

 

Yeah....I don't use Asus turbo anything...ever. Regardless of the board maker, everything eventually ends up manually tuned from fans to CPU to memory.

 

 

 

[jaybee83]

1 hour ago, cylix said:

AMD Ryzen 7 9800X3D Overclocked to 5.46 GHz, Beating Ryzen 7 7800X3D by 24%

while those are some promising numbers, WHY THE HECK DO THEY ALWAYS CHOOSE GEEKBENCH?! that bench is sooooo completely useless 😄 

[Tenoroon]

1 hour ago, Mr. Fox said:

It is awesome to see that kind of performance jump. Anyone that doesn't like that is either lying or stupid.

 

Speaking of stupidity, I think one of AMD's biggest challenges is the mentality of some of their customers. Like this one... That is detrimental to all concerned, include those that don't care but are too dumb to know better.

 

I think it is inconclusive to judge this man on his statement. He does not specify if he cares about overclocking existing.

Here's an example. I dislike EV's, they are conceptually very interesting, but I believe that they aren't much better than ICE vehicles in anything, and are just generally pricey. However, as much as I dislike EV's, if all EV's were to get banned, I would be upset because there are people that care about them/like them, and for them to have what they enjoy taken is unfair in my opinion.

I think it is reasonable to state that one does not care about overclocking, while I disagree with the statement, JohN, in this instance, is just stating how he feels. Never does he mention his thoughts on if OC is something that should be available for all consumers. If he had noted that overclocking is dumb and no one should do it or be able to do it, then I think it would be fair to argue against his point and try to show him why his thinking is flawed.

Now, I do see why this looks bad. Any AMD exec could read this and assume that because JohN does not care about OC, that his statement is implying that he would be impartial or happy if OC on AMD processors was permanently disabled. I believe that, in this scenario, the person to blame is the one making an assumption—in this case, the imaginary AMD exec—as that assumption could have been easily avoided with more inquiry into JohN's thoughts on OC, particularly whether he believes it should exist at all. If I were JohN, and the aforementioned assumption was made, I would be displeased because I would feel that the AMD exec took what I said completely out of context.

I'm not trying to start an argument or get political here, I just disagree with judging this man on his statement. If @Mr. Fox or anyone else disagrees with me, we can agree to disagree. One of the best things that is hard to acknowledge sometimes is that we are blessed enough to live in a society where we can freely agree and disagree without facing any kind of persecution.

[tps3443]

Stability testing 192 Gigs of ram is very time consuming fellas lol. 
 

So, when I ran my 2x24GB (48GB) DDR5 kit I could fine tune voltages and timings and do a quick check in TM5 1usmus which would take 12-15 minutes. And then I could move to longer stability testing.
 

That same 12-15 minutes of 1usmus now takes 2+hours LOL. 
 

So, if someone tells me (3cycles) 1usmus ain’t good enough. I’m gonna be like BRUH seriously….🤣

[Mr. Fox]

3 hours ago, jaybee83 said:

while those are some promising numbers, WHY THE HECK DO THEY ALWAYS CHOOSE GEEKBENCH?! that bench is sooooo completely useless 😄 

Probably because it is a benchmark that Ryzen shines in. The opposite of what it would look like with AIDA64 Cache and Memory Benchmark. If you look on HWBOT we see really good numbers from Ryzen in Geekbench and it includes memory performance testing, just from a different angle than AIDA64, but it is not categorized as a memory benchmark. It is considered a CPU benchmark, but it is affected by memory performance.

2 hours ago, Tenoroon said:

I think it is inconclusive to judge this man on his statement. He does not specify if he cares about overclocking existing.

Here's an example. I dislike EV's, they are conceptually very interesting, but I believe that they aren't much better than ICE vehicles in anything, and are just generally pricey. However, as much as I dislike EV's, if all EV's were to get banned, I would be upset because there are people that care about them/like them, and for them to have what they enjoy taken is unfair in my opinion.

I think it is reasonable to state that one does not care about overclocking, while I disagree with the statement, JohN, in this instance, is just stating how he feels. Never does he mention his thoughts on if OC is something that should be available for all consumers. If he had noted that overclocking is dumb and no one should do it or be able to do it, then I think it would be fair to argue against his point and try to show him why his thinking is flawed.

Now, I do see why this looks bad. Any AMD exec could read this and assume that because JohN does not care about OC, that his statement is implying that he would be impartial or happy if OC on AMD processors was permanently disabled. I believe that, in this scenario, the person to blame is the one making an assumption—in this case, the imaginary AMD exec—as that assumption could have been easily avoided with more inquiry into JohN's thoughts on OC, particularly whether he believes it should exist at all. If I were JohN, and the aforementioned assumption was made, I would be displeased because I would feel that the AMD exec took what I said completely out of context.

I'm not trying to start an argument or get political here, I just disagree with judging this man on his statement. If @Mr. Fox or anyone else disagrees with me, we can agree to disagree. One of the best things that is hard to acknowledge sometimes is that we are blessed enough to live in a society where we can freely agree and disagree without facing any kind of persecution.

You are not wrong. He is certainly entitled to have an opinion, but sometimes expressing opinions in the wrong context can only be viewed as trolling or inappropriate. The issue is he was making comments on a article specifically about overclocking, and celebrating those results, with a comment he does not care about overclocking. The article is for those that do care.

 

Kind of like going to a pizzeria and declaring you don't like pizza, but it's nice that you serve pizza for those that do. That person's presence at the pizzeria benefits no one and the comments serve no purpose. That person's motives for being there and his intelligence in making such a comment at that time and place might be legitimately called into question. That isn't persecution, it is expressing an opinion or making a statement that isn't relevant or useful and puzzling to those that do like pizza. The question one might ask is "Then why are you here?"

[Talon]

https://ibb.co/VBdv7db

 

Stock 250/250 and 347a power and current limits. Pulls a max of 240w during this run. Not too shabby for a such a low power limit and only 24 threads. I might go direct die on this chip. Seems that EK Direct Die AIO is compatible with 1851 according to their website. 

[electrosoft]

47 minutes ago, Talon said:

https://ibb.co/VBdv7db

 

Stock 250/250 and 347a power and current limits. Pulls a max of 240w during this run. Not too shabby for a such a low power limit and only 24 threads. I might go direct die on this chip. Seems that EK Direct Die AIO is compatible with 1851 according to their website. 

 

Nice! I will say Provantage was very easy to cancel with and I had not one but TWO CSRs reach out to me to cancel my order. I'm definitely keeping an eye on it and seeing all you can extract from it bro!

 

So far, how would you say it compares to your 13th/14th gen setup?

 

 

[Mr. Fox]

3 hours ago, electrosoft said:

 

Nice! I will say Provantage was very easy to cancel with and I had not one but TWO CSRs reach out to me to cancel my order. I'm definitely keeping an eye on it and seeing all you can extract from it bro!

 

So far, how would you say it compares to your 13th/14th gen setup?

 

 

Yes, I have purchased from them a couple of times and they are excellent. I even had their head of sales call me to apologize for a mistake they had made. It wasn't a big deal in the grand scheme of things, but the fact that they cared enough was pretty great. We don't see that often in the tech space at all, especially with EVGA out of the picture.

[Papusan]

 

It seems Arrow Lake Mobile won't be any better than the new released and flopped Core Ultra desktop chips. Better power efficiency who is Intel's main goal and target. Oh'well, that's as expected, but people get something brand new so maybe they are still happy with what they get. 

 

Intel Core Ultra 9 285H makes Geekbench debut with 16-core CPU and severe throttling

 

While the performance will undoubtedly be better at launch, the Core Ultra 9 285H doesn't seem like that much of an improvement over Meteor Lake. Hopefully, it compensated by an increase in power efficiency, something that has been on Intel's crosshairs since Lunar Lake launched earlier this year.

 

 

Intel has a laptop problem: its Core Ultra 9 285H is half as fast as the PC 285K

Intel keeps its promise of bringing a new generation of processors every year, offering a series of CPU models that should improve on the previous ones. Sometimes the improvement is minimal and not even worth it if it costs more,...

[Mr. Fox]

2 hours ago, Papusan said:

 

It seems Arrow Lake Mobile won't be any better than the new released and flopped Core Ultra desktop chips. Better power efficiency who is Intel's main goal and target. Oh'well, that's as expected, but people get something brand new so maybe they are still happy with what they get. 

 

Intel Core Ultra 9 285H makes Geekbench debut with 16-core CPU and severe throttling

 

While the performance will undoubtedly be better at launch, the Core Ultra 9 285H doesn't seem like that much of an improvement over Meteor Lake. Hopefully, it compensated by an increase in power efficiency, something that has been on Intel's crosshairs since Lunar Lake launched earlier this year.

Intel has a laptop problem: its Core Ultra 9 285H is half as fast as the PC 285K

Intel keeps its promise of bringing a new generation of processors every year, offering a series of CPU models that should improve on the previous ones. Sometimes the improvement is minimal and not even worth it if it costs more,...

I wish I could say that I have a surpised look on my face. Even if it had turned out well on the desktop side of things it would have been taken for granted that the mobile side of things would suck because that is always the case for mobile. The fact that this did not turn out well on the desktop made the prospect of a good mobile product far more unlikely. 😑

[Mr. Fox]

On 10/19/2024 at 11:15 AM, Mr. Fox said:

I came very close to buying new AMD parts the other day, but then I pulled my head out and remembered how that has turned out every time before and that I have never been satisfied, even once, with anything they make. It's also an engineering abortion with a tiled design and a guaranteed crappy CCD paired with a potentially good or OK CCD. But, the fact that I even paused to momentarily consider doing that shows how disappointing the newer and not-better platform is.

On 10/19/2024 at 11:27 PM, electrosoft said:

Oh man, this thread would have been filled with some serious fire and brimstone from you if you made that purchase!

Well, stay tuned. There may be some hell to pay.

[Clamibot]

15 minutes ago, Mr. Fox said:

Well, stay tuned. There may be some hell to pay.

We are on page 666 after all.

[win32asmguy]

2 hours ago, Papusan said:

 

It seems Arrow Lake Mobile won't be any better than the new released and flopped Core Ultra desktop chips. Better power efficiency who is Intel's main goal and target. Oh'well, that's as expected, but people get something brand new so maybe they are still happy with what they get. 

 

Intel Core Ultra 9 285H makes Geekbench debut with 16-core CPU and severe throttling

 

While the performance will undoubtedly be better at launch, the Core Ultra 9 285H doesn't seem like that much of an improvement over Meteor Lake. Hopefully, it compensated by an increase in power efficiency, something that has been on Intel's crosshairs since Lunar Lake launched earlier this year.

 

 

Intel has a laptop problem: its Core Ultra 9 285H is half as fast as the PC 285K

Intel keeps its promise of bringing a new generation of processors every year, offering a series of CPU models that should improve on the previous ones. Sometimes the improvement is minimal and not even worth it if it costs more,...

 

This chip (285H) does not make sense as OEMs are not going to magically reverse course and provide sufficient cooling to sustain advertised clocks and performance. If its paired with any kind of mobile dedicated GPU expect performance to drop even more. Of course this line will also be a locked chip so you cannot properly tune it to be more efficient.