Some Intel Nova Lake CPUs Rumored to Challenge AMD's 3D V-Cache in Desktop Gaming

[Hecate91]

I am always buying more AMD boards in the long run than Intel..why is that?

I'm not sure why besides FOMO. If you have an X670 board you're going to have Zen 6 support.

Because even Intel's old RPL is still competitive to AMD's what? 7900x, 7950x, 7800X3D, 7950X3D, 7900X3D, 9950X, 9900X, 9800X3D, 9900X3D and 9950X3D..its just 1 intel chip for what? 8 AMD chips..

Raptor lake was 2 generations of chips, well sort of if you want to count the refresh. AMD Zen 4 & Zen 5 are 2 chip generations as well.
But all of the Raptor Lake failures probably turned away quite a few users, given how Intel handled it.

 

[JustBenching]

Raptor lake was 2 generations of chips, well sort of if you want to count the refresh. AMD Zen 4 & Zen 5 are 2 chip generations as well.
But all of the Raptor Lake failures probably turned away quite a few users, given how Intel handled it.

Replace all faulty parts and extended warranty by an extra 2 years for everyone. Yes, terrible handling. They should take lessons from amd with their 3d chips popping like cherries for the last 2 years

 

[Hecate91]

Replace all faulty parts and extended warranty by an extra 2 years for everyone. Yes, terrible handling. They should take lessons from amd with their 3d chips popping like cherries for the last 2 years

An extra 2 years on parts which could still fail, after months of Intel throwing the blame at their customers and claiming a BIOS update would fix a hardware defect.
I'm not surprised you bring up the X3D issue, which was only a few cpu's in comparison, and was the fault of mobo manufacturers pushing up voltages.

 

[JustBenching]

An extra 2 years on parts which could still fail, after months of Intel throwing the blame at their customers and claiming a BIOS update would fix a hardware defect.
I'm not surprised you bring up the X3D issue, which was only a few cpu's in comparison, and was the fault of mobo manufacturers pushing up voltages.

Yes yes, im telling you, id feel a lot better with my x3d knowing it's a mobo issue and im not getting any extra warranty

Even if put your feet to the fire you won't admit the obvious, intel handled it impeccably good.

 

[A Computer Guy]

That's like saying you can't compare efficiency of cars by running them at the same speed.

To be more analogous to my reply I'm saying you don't typically compare fuel efficiency of cars by running them at the same horsepower. In the US anyway it's Miles per Gallon. In computing terms a useful CPU efficiency measurement is work done by time. To phrase it differently I'm going to borrow some reddit text that expresses my point better as I understand it. "Efficiency is how much work a CPU can do with a certain amount of power. So a 1000W CPU could be very efficient by comparison if it can do 20x the work of a 100W CPU in the same amount of time."

If you need to run within a certain power budget then a fixed wattage changes the efficiency outcome by that specific scenario. From arguments back and forth on TPU I get the idea that Intel (12/13/14th gen+) is more efficient at lower half of Ryzens' power curve but Ryzen is more efficient toward the upper half of it's power curve. (talking about multi-core)

To bring this point closer to the topic of v-cache, if Intel decides not to bring it's v-cache solution to desktop then AMD will likely keep the crown of efficiency at least among games that benefit from v-cache.

 

[RootinTootinPootin]

But all of the Raptor Lake failures probably turned away quite a few users

"few"

Intel "Did" something back then by offering extended RMA's, and its widely caused by setting every single voltage setting to "Auto", me myself and I never used Auto, my chips are still alive and stomping at 6ghz all day.. (2x13900KS,1x14700K,1x13600K)

I'd never be turned off by such a minor (spun to major) issue which I believe is easy to solve when you learn to read, rather trusting YTube for everything, even food..

AMD on the other hand, denies any warranty for both board and processor when it catches fire..my 9950X3D belly bottom is having a change of color, I dunno why, Asus never helped, nor AMD, and I believe I will be left alone somewhere when it burns..

 

[JustBenching]

To be more analogous to my reply I'm saying you don't typically compare fuel efficiency of cars by running them at the same horsepower. In the US anyway it's Miles per Gallon. In computing terms a useful CPU efficiency measurement is work done by time. To phrase it differently I'm going to borrow some reddit text that expresses my point better as I understand it. "Efficiency is how much work a CPU can do with a certain amount of power. So a 1000W CPU could be very efficient by comparison if it can do 20x the work of a 100W CPU in the same amount of time."

If you need to run within a certain power budget then a fixed wattage changes the efficiency outcome by that specific scenario. From arguments back and forth on TPU I get the idea that Intel (12/13/14th gen+) is more efficient at lower half of Ryzens' power curve but Ryzen is more efficient toward the upper half of it's power curve. (talking about multi-core)

But miles per gallon doesn't say much unless the cars are running at the same speed, right? If car A runs at 200mph and car B at 50mph car B will look more efficient, even thought it might not be.

The same analogy with cpus would either be both running at the same power or both running at the same "speed", and speed would be time to finish the task. For example how much power does a 265k need to get 30k CBR score vs how much power does the 9700x need? Im sure the amd part will consume many times more power than the intel part. The arguments back and forth have convinced you of that because they are comparing intel unlocked parts that draw a gazilion power out of the box. If you limit them to the same power as their amd counterparts, OR test the actual locked parts (non k / T cpus) youll realize that especially in the midrange due to the core count deficiency AMD chips are laughably bad in the efficiency department.

 

[A Computer Guy]

But miles per gallon doesn't say much unless the cars are running at the same speed, right? If car A runs at 200mph and car B at 50mph car B will look more efficient, even thought it might not be.

I think you are coming around to being on the same page with me now. Here's where it probably also matters to clearly state what kind of efficiency you are looking for. If time-efficiency is what you want to measure then you need to run each car at it's fastest speed. If fuel-efficiency or not getting a traffic ticket efficiency is what you are after then you probably want to run them at a comparable speed like the speed limit.

CPU's aren't cars and the ramification of locking frequency directly can drastically effect power management so you have to keep in mind to a certain extent you are comparing apples to oranges when frequency is not limited. In my opinion anyway the efficiency metric that seems to matter the most and easiest to compare is being able to get the most work done in a fixed amount of time. When it comes to comparing CPU's it's nice to see comparisons both out of the box and with reasonable power levels to get a more complete picture comparing efficiency knowing that it's not a perfect comparison unlike comparing a K vs. non-K part that are architecturally the same.

When it comes to potential v-cache Intel enabled parts many will be looking at FPS figures as their metric of comparison for games that rely on v-cache.

The same analogy with cpus would either be both running at the same power or both running at the same "speed", and speed would be time to finish the task. For example how much power does a 265k need to get 30k CBR score vs how much power does the 9700x need? Im sure the amd part will consume many times more power than the intel part.

Speed is speed, time is time, your mixing things up here.

The arguments back and forth have convinced you of that because they are comparing intel unlocked parts that draw a gazilion power out of the box.

To my understanding Intel advocates say running K parts at non-K power levels is essentially the same thing as a non-K part.

If you limit them to the same power as their amd counterparts, OR test the actual locked parts (non k / T cpus) youll realize that especially in the midrange due to the core count deficiency AMD chips are laughably bad in the efficiency department.

Did you just contradict yourself? "Just Benching previously said: That's why if you eg compare a 265k to a 265 non k the latter will look a lot more efficient, because it has a lower power limit, but in reality it isn't actually more efficient. If anything it might even be worse due to worse binning." Your non-K comments just entered schrodinger's cat territory I think.

 

[JustBenching]

Did you just contradict yourself? "Just Benching previously said: That's why if you eg compare a 265k to a 265 non k the latter will look a lot more efficient, because it has a lower power limit, but in reality it isn't actually more efficient. If anything it might even be worse due to worse binning." Your non-K comments just entered schrodinger's cat territory I think.

How is that a contradiction? Im saying if you don't want to touch the power limits on the k parts then get a non k part and leave it as is out of the box. Nothing contradictory about it.

You grab a 9700x and a 265k, you lock them to the same power or target the same performance (eg 30k score in cbr), the 265k will both be faster and more efficient. I don't see why we are exchanging 30 posts about something so trivial. The 9700x needs 600 watts to score 30k, the 265k need 150.

 

[A Computer Guy]

How is that a contradiction? Im saying if you don't want to touch the power limits on the k parts then get a non k part and leave it as is out of the box. Nothing contradictory about it.

Perhaps I misunderstood something.

You grab a 9700x and a 265k, you lock them to the same power or target the same performance (eg 30k score in cbr), the 265k will both be faster and more efficient.

If you lock the cpus to a specific CB score you probably arn't running each individual cpu at their peak efficiency making a comparison of CPU efficiency pointless.
At best your getting a statistic on how much power will be consumed at a given CB score which might not be a realistic representation of efficiency for either CPU.

I don't see why we are exchanging 30 posts about something so trivial.

It's Sunday.

The 9700x needs 600 watts to score 30k, the 265k need 150.

I don't think running a 9700x at 600 watts is an option so what you just said is irrelevant and non-sensical.

I missed the fact your comparing a 20 core part to an 8 core part.
What a waste of time per word efficiency!
My v-cache is full and this is where I am going to end.

 

[blinnbanir]

I can see this thread getting locked again. Intel is in the rear view in DIY already. As teenagers who have been with AMD since 2017 will lead to an increase in AMD market share. People who are ragging on AMD can thank them for 6 and 8 core CPUs. If Intel were in the lead you would still be getting 4 core CPUs in the Consumer market. Then there is the truth. By the time Nova Lake launches AMD will have something to respond. I am amazed at how much faster the 9900X3D is vs the 7900X3D and that was already a fast chip. One thing it proves is Intel have to respond to X3D as it was them that started the narrative in DIY when the 1700X launched with 8 cores all those years ago. The best though I won't need to change my MB.

 

[JustBenching]

If you lock the cpus to a specific CB score you probably arn't running each individual cpu at their peak efficiency making a comparison of CPU efficiency pointless.
At best your getting a statistic on how much power will be consumed at a given CB score which might not be a realistic representation of efficiency for either CPU.

Is there any cb score that the 9700x wins? Let's say at 20k, do you reckon it stands any kind of a chance?

I missed the fact your comparing a 20 core part to an 8 core part.

Exactly. That's why the intel chips are stupidly more efficient. Cause they are not on a core deficiency diet like their amd counterparts.

 

[Tyler-98-W68]

Intel had to respond to the 1700x, when their existing 4 core cpus were better in almost eveything due to the superior IPC?

 

[blinnbanir]

Is there any cb score that the 9700x wins? Let's say at 20k, do you reckon it stands any kind of a chance?


Exactly. That's why the intel chips are stupidly more efficient. Cause they are not on a core deficiency diet like their amd counterparts.

What an out of touch Comment. AMD killed Intel with cores and the Gaming industry is introducing Games to the point where 4 cores have become irrelevant in modern Games. I guess you are not banned for entertainment purposes though.

Intel had to respond to the 1700x, when their existing 4 core cpus were better in almost eveything due to the superior IPC?

IPC had nothing to do with Copying DVDs and more cores meant faster copying. There is a reason the DVD rental industry died. Especially when it came to Games and Anime. 2017 also does not matter in 2025. IPC was gained with the 2700X and the 3000 chips were so good that they slowed down 5000 sales until the 5800X3D was launched and 5900 chips fell in price. There is no doubt that EUV from TSMC vs what Intel was using, as their hubris made them feel they did not need EUV has meant that in 2025, combined with AMD's socket support make them objectively better than Intel to the point where Distributors charge a premium to retailers for those parts and MB makers charge super premiums for MBs to the point where the most expensive AM5 MB is double a decent TRX50 and has 1/4 the bandwidth.

 

[JustBenching]

Just so we clear, intel had been offering 6 core parts for 390$ since 2013. Just so we are more clear, amd has been offering 6 and 8 core parts for the last 7 years in the <450$ range. Pretending that amd killed intel with core counts is completely delusional, as delusional as getting 150 fps in spiderman maxed out

I don't get how teh same people complaining about intel's 4 core 10 years ago are fine with 6 core and 8 core parts from amd when their competition offers 20. Brand fanboyism is the death of reason

 

[Tyler-98-W68]

IPC doesn't matter, cope harder. So IPC didn't matter during the bulldozer days? Ok let me rephrase that. Fast single core performance doesn't matter?

A game might be slightly multithreaded, but it doesn't want 16 cores of POS slow cores. Tell me who has the fastest single threaded performance right now?

But im sure you will tell me that doesn't matter either right?

 

[A Computer Guy]

Is there any cb score that the 9700x wins? Let's say at 20k, do you reckon it stands any kind of a chance?

I hate to continue commenting on this but disable the ecores so you are comparing 8c vs 8c then have a go at it if you really want to know how things land.

Exactly. That's why the intel chips are stupidly more efficient. Cause they are not on a core deficiency diet like their amd counterparts.

Given both CPU's are modern the 20 cores will get more work done in a given time but also consumes more power than 8. 9700x is more power efficient overall while the 265k gets more done overall (where more cores count) but is less power efficient in doing so (because more cores more power draw). I'll leave the mental exercise to you. If you doubt me then just look at all the benchmark charts doing the comparisons and explain why they are wrong.

 

[JustBenching]

I hate to continue commenting on this but disable the ecores so you are comparing 8c vs 8c then have a go at it if you really want to know how things land.

Why would you do that?

Given both CPU's are modern the 20 cores will get more work done in a given time but also consumes more power than 8.

They will consume more power if you allow them to consume more power. If you allow them to consume the same power they will sitll get more work done and end up with less overall power usage since they will finish faster. Hence, they are more efficient.

And i followed your advice, looked at the reviews. The 265k at 95w scores 28k in CBR23. Explain to me how the 9700x is more efficient than that

Intel Core Ultra 7 265K cooling testing: How much does it take to keep Arrow Lake cool in MSI's MPG Velox 100R PC Case?

Intel’s Core Ultra 7 265K is easier to cool than previous-generation CPUs like the Core i7-14700K.

www.tomshardware.com

 

[Tyler-98-W68]

Now we are talking about disabling e-cores to make it fair? Can we disable 3dcache too?

 

[Panther_Seraphin]

It would be interesting if Intel reentered HEDT with their version of 3D v-cache. Threadripper needs some competition to balance pricing. And this makes me question where the heck is X3D for Threadripper?

I doubt it will come as it will eat into EPYC sales unfortunately for us consumers.

 

[A Computer Guy]

Why would you do that?

You know why.

They will consume more power if you allow them to consume more power. If you allow them to consume the same power they will sitll get more work done and end up with less overall power usage since they will finish faster. Hence, they are more efficient.

Then prove it and post back here (in different thread called "Ha Ha, I told you so...")

 

[JustBenching]

You know why.

Then prove it and post back here (in different thread called "Ha Ha, I told you so...")

Sure, here you go. 28k at 95w, the 9700x barely scores 22

Intel Core Ultra 7 265K cooling testing: How much does it take to keep Arrow Lake cool in MSI's MPG Velox 100R PC Case?

Intel’s Core Ultra 7 265K is easier to cool than previous-generation CPUs like the Core i7-14700K.

www.tomshardware.com

 

[Panther_Seraphin]

Just so we clear, intel had been offering 6 core parts for 390$ since 2013. Just so we are more clear, amd has been offering 6 and 8 core parts for the last 7 years in the <450$ range. Pretending that amd killed intel with core counts is completely delusional, as delusional as getting 150 fps in spiderman maxed out

I don't get how teh same people complaining about intel's 4 core 10 years ago are fine with 6 core and 8 core parts from amd when their competition offers 20. Brand fanboyism is the death of reason

AMD was offering 6 core processors for <$300 back in 2010.


AMD killed Intel with a processor with limited compromises. It could keep up/beat HEDT offerings in productivity without the HEDT Tax while being within single digit percentages of the desktop class processors in gaming without requiring a nuclear power stations cooling or output to keep up.

 

[Hecate91]

Intel had to respond to the 1700x, when their existing 4 core cpus were better in almost eveything due to the superior IPC?

Intel responded with a 6 core right after AMD launched Ryzen, after years of stagnating the market with 4 cores. And no Intel HEDT doesn't count as pricing was out of reach for the average person building a gaming pc.

IPC doesn't matter, cope harder. So IPC didn't matter during the bulldozer days? Ok let me rephrase that. Fast single core performance doesn't matter?

A game might be slightly multithreaded, but it doesn't want 16 cores of POS slow cores. Tell me who has the fastest single threaded performance right now?

But im sure you will tell me that doesn't matter either right?

IPC doesn't matter as much as multicore and performance per watt does. Arrow lake has less IPC and is less efficient than Raptor lake. But it's interesting how the Intel fans claimed multithread didn't matter because it made their 4 cores look bad, now thanks to AMD making 8 cores mainstream, multithread performance is important.

 

[JustBenching]

AMD was offering 6 core processors for <$300 back in 2010.


AMD killed Intel with a processor with limited compromises. It could keep up/beat HEDT offerings in productivity without the HEDT Tax while being within single digit percentages of the desktop class processors in gaming without requiring a nuclear power stations cooling or output to keep up.

Delusional take. Until 2022 with thr first x3d chips amd was behind not just in gaming performance but also in gaming power draw. Amd fans....