• 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.

14th Gen "Raptor Lake Refresh"

Yes true. All Raptor Lake is doing is adding more e-waste cores. And the p-cores have almost no IPC improvement. Zen 4 looks like a Zen 3 refresh with improved clocks and very little IPC improvement. Golden Cove already has 15-20% better IPC at same clock speed across most workloads compared Zen 3 so Zen 4 will still be behind Golden Cove IPC with only an 8% IPC uplift.

Zen 4 though does bring exciting platform improvements like PCI-E Gen 5 on whole chipset of X670E. Though I do not care about that but those who desire Gen 5 PCI-E SSDs it could be great.

With Raptor Lake, they could be monster gaming chips with the added L3 cache and be what the 5800X3D was to 5800X when you shut off e-waste cores. They may be the best 8 core powerhouse chips around with ability to super clock 8 P cores the ring so high with 40-45MB L3 cache instead of 25-30MB on 8 P core Alder Lake counterparts. I bet that gives a big boost to gaming when shutting down e-waste cores and you have the best 8 core 16 thread productivity chip with a Core i7 13700K.

But Zen 5 is going to bring 16 cores per CCD so first chip with modern arch with more than 8 cores on one CCD/ring for no high latency penalty to cross. That could be big especially if you want more than 8 cores on same CCD/ring with no cross latency penalty.

Meteor Lake who knows what it is going to be. t looked like an 8+24, then they changed to 3 different cores types. They may keep changing it. Hopefully more than 8 super fast P cores or at least 8 P cores with a super high clocks and IPC uplift of 50% or more.
"e waste cores". I just tested FC6 and it works way better with e core on rather than off. More performance and less power draw at the same time. Do you have an alderlake?
 
"e waste cores". I just tested FC6 and it works way better with e core on rather than off. More performance and less power draw at the same time. Do you have an alderlake?

Yes have a 12700K and oh my the p cores are so awesome.

Manually tuning it provides less power raw than out of the box where power goes through roof as it overvolts way more than necessary.


Do you have lots of stuff runing in background that happens to not get confused with the e-cores on WIN11. Then it may help.

If you do not runs stuff in background, very few games use more than 6 cores let alone 8, so without a lot of stuff like streaming and such running in background, having e-cores off all else being equal games will run better even Hitman 3 which is one of the only games to offload the audio engine to e-cores.

Do you have it on auto. If you disable e-cores power usage goes up as auto settings I believe try and shove lots more volts and frequency through P cores causing more power usage as P cores take much more power than even a 4 e-core cluster.

I call them e-waste cores as I do not like them and feel they belong in mobile and tablet devices only and they drag the ring clock down, but love the p cores which is why Alder Lake is so good. To me it is so good anyways as long as you shut off the e-cores you get by far the best IPC and clocked 8 core chip in existence 15-20% better IPC than Zen 3 and can clock so much higher and easier especially the ring with e-cores off.
 
MTL would be awesome if they create P only versions, erase those e-craps and give us 10-12 P cores, it would be brutal for gaming, and even for content creation workload, it is 2-4 more P cores than current, and each P core is much more powerful than crap e-core so they don't need as much as e-cores, and you don't need specific optimization to not let game use e-core which decrease performance.
 
MTL would be awesome if they create P only versions, erase those e-craps and give us 10-12 P cores, it would be brutal for gaming, and even for content creation workload, it is 2-4 more P cores than current, and each P core is much more powerful than crap e-core so they don't need as much as e-cores, and you don't need specific optimization to not let game use e-core which decrease performance.

Bingo I could not agree more. Hate those e-waste cores. Love the P cores so much. The P cores rock and are so much better than even AMD's best Zen 3 right now. But e-waste cores I hate them and they are a waste of silicon IMHO. Fortunately for now, 8 cores is more than enough for gaming which Intel provides the best of. Though we need more than 8 P cores on Meteor Lake as an option as eventually games may use more than 8 cores though probably not for at least a few years or longer?? Plus even if games do not use more than 8 good cores, more P cores is good for better running background tasks and streaming and such while gaming without the gimmick hybrid crap of e-waste cores on desktop (they belong in mobile devices) and scheduling issues plus risk of an important tasks getting assigned to e-waste core which will cripple performance.
 

Intel confirms 14th Gen Meteor Lake has ‘Versatile Processing Unit’ for AI/Deep Learning applications​

Source: https://videocardz.com/newz/intel-c...essing-unit-for-ai-deep-learning-applications

Tm9bRmk.jpg

Videocards said:
Intel is adding VPU to Meteor Lake and newer.

A new commit to Linux VPU driver today confirms that the company has plans to introduce a new processing unit into consumer 14th Gen Core processors, a Versatile Processing Unit.

The VPU driver is included into the Linux Direct Rendering Manager (DRM), the same way their graphics driver is integrated. The VPU appears 6 years after Intel acquired a company called Movidius, which has been developing their own VPUs. It is not entirely clear if and how Intel plans to incorporate Movidius designs into Meteor Lake, it could be a full-blown SoC-like integration or just a copy of architecture bits needed for Meteor Lake. Obviously after so many years, VPU design should be much more complex.

Thus, Intel confirms the new VPU has five components, including CPU to VPU integration unit, memory management, RISC controller, network on chip and the most important part, the Neural Compute Subsystem (NCS) doing the actual work. This VPU unit could be considered Intel’s alternative to NVIDIA’s Tensor Cores, a dedicated chip that is heavily focused on AI algorithms.

Intel Meteor Lake is now officially coming next year, eventually it should become available for mobile and desktop platforms packed with new hybrid architecture featuring Redwood Cove and Crestmont CPU cores and Intel’s newest Xe-LPG graphics architecture.

Hoping to see some performance leaks before Christmas!
 
Looks like Meteor Lake is now delayed until the end of 2023 at the earliest, so maybe no launch until 2024.

 
Looks like Meteor Lake is now delayed until the end of 2023 at the earliest, so maybe no launch until 2024.

I think Intels production issues will cost them ultimately. HAving teir own fabs is seemingly not as beneficial as was previously discussed
 
I think Intels production issues will cost them ultimately. HAving teir own fabs is seemingly not as beneficial as was previously discussed

The issue seems to be irrelevant of the node or Fab

The trend force article is written weird but what I'm reading on others is they the issue is the GPU itself and Intel needs to delay the entire product due to issues with the GPU. They had asked Tsmc to make the GPU

Right now it looks like Intels entire graphics division is a right old mess
 
Right now it looks like Intels entire graphics division is a right old mess

You've misinterpreted the article details, the slowdown/cancellation of the 3nm TSMC parts for MTL is what is indicating the delay of the actual CPU not that the part being made by them that is the issue. Seems that a larger issue with the product for them to cancel mass production by a further 2-3 quarters, but I can see why they wouldn't want to be sitting on a huge pile of 3nm graphics tiles if the rest of the CPU is being delayed.
 

Intel Meteor Lake iGPU might support ray tracing acceleration​


Videocardz said:
A few weeks ago Intel confirmed their upcoming 14th Gen Core series are to feature “Versatile Processing Unit” for AI and inference. Now Intel development team is revealing some details on the integrated graphics architecture.

Source: https://videocardz.com/newz/intel-meteor-lake-igpu-might-support-ray-tracing-acceleration


Hopefully this won't use more % of the CPU's die/package space that what's used currently. With AMD also sacrificing package space to include a iGPU for Zen4 I guess it's not so much of an issue, as this was a good advantage AMD had over Intel for the last few years.
 

Meteor Lake CPUs Could Play Videos, Use QuickSync Without a GPU​


Source: https://www.tomshardware.com/news/meteor-lake-plays-videos-without-gpu

Tomshardware said:
In a surprising twist, Intel's future 14th Generation Meteor Lake CPU architecture will reportedly split video playback and encode functionality from the integrated graphics into a new area known as an SMU or Standalone Media Unit on the CPU, as reported by Phoronix. It is a significant change by Intel, which will allow media functionality to be used at all times, even if the integrated graphics chip is disabled.

Pretty interesting. Hope we get a clear date for Meteor Lake launch in the coming months. Be nice to see Intel on 7nm (Intel 4 as it'll be named) - wonder how this competes with TSMC 5nm in performance/yield/power efficiency.
 
They've always said 2023 for MTL developed with their latest 7nm process ('Intel 4').

Bare in mind that their 7nm process CPUs were originally meant to ready around 2020, so I think 3 years of delays would be a bit of a stretch.

EDIT - it's quite possible that mobile chips will be in 2023, desktop in 2024 H1, judging by the fact that Intel have said nothing yet about a desktop series in 2023.

That would likely mean that Meteor Lake on DDR5 only, will be going up against Zen 5 in 2024, but Intel will release their series at least 3 months earlier than AMD.

I really hope they don't make the mistake of trying to cram in more E-cores than P-Cores like with Raptor Lake. In my opinion, they should limit E-cores to a 1:1 ratio on desktop CPUs.

I wonder if the desktop CPUs will be fabbed on 'Intel 3' only, due for volume production in the first half of 2024?

AnandTechRoadmaps3.png
 
Last edited:
Back
Top Bottom