Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads

[Tigerick]

As Hot Chips 34 starting this week, Intel will unveil technical information of upcoming Meteor Lake (MTL) and Arrow Lake (ARL), new generation platform after Raptor Lake. Both MTL and ARL represent new direction which Intel will move to multiple chiplets and combine as one SoC platform.

MTL also represents new compute tile that based on Intel 4 process which is based on EUV lithography, a first from Intel. Intel expects to ship MTL mobile SoC in 2023.

ARL will come after MTL so Intel should be shipping it in 2024, that is what Intel roadmap is telling us. ARL compute tile will be manufactured by Intel 20A process, a first from Intel to use GAA transistors called RibbonFET.





Intel Core Ultra 100 - Meteor Lake



As mentioned by Tomshardware, TSMC will manufacture the I/O, SoC, and GPU tiles. That means Intel will manufacture only the CPU and Foveros tiles. (Notably, Intel calls the I/O tile an 'I/O Expander,' hence the IOE moniker.)

 

[DrMrLordX]

"Testing regime that has been used for mobile CPUs for years" - testing regime that is basically Cinebench on loop?

If that's what laptop review sites used previously, then yes. And nobody really seemed to complain overmuch until now.

 

[Abwx]

Yup.

Two distinct possibilities as to why the early review samples had months old pcode. First is that's the pcode version manufacturers were originally supplied with when starting their product validation and no one told them that they needed to update before sending out the review samples. Second is that it was intentional - get one mixed review cycle on launch, then get a second thanks to this large efficiency improvement as more laptop designs become available.

There s a third possibility, and it s the actual one from what i did read, this guy is a chinese intel employee...

Besides he use old drivers that preceded the launch drivers, so much for the last minute pcode revision..

 

[mikk]

There s a third possibility, and it s the actual one from what i did read, this guy is a chinese intel employee...

I read somewhere he worked or still works for Lenovo.

Besides he use old drivers that preceded the launch drivers, so much for the last minute pcode revision..

Not sure what you mean. He tested an early enabling model and on these models the driver and firmwares are older than it could have been. Note that Intel didn't supply anyone with reviewer/reference devices and therefore almost everyone did use the pre installed old driver, it's not like Intel supplied new drivers to these testers. Most didn't even have a production unit, mostly pre-production units. Ultrabookreview, hardwareluxx, Notebookcheck, Computerbase all tested with a pre-production unit.

 

[Geddagod]

If the pcode revision stuff is accurate, I think it's pretty easy to explain.
In order for OEMs to enable/run all the chips as part of their testing and for ES/QS samples, Intel's predefined V/F curve was extremely loose. This ensured that even ES samples which might need the extra voltage to hit clocks and run get that extra voltage needed.

 

[mikk]

Ultrabookreview tested a Zenbook with Bios 201, Asus support has bios updates 202 and 203 on their website with "Optimize system performance"

 

[Abwx]

I read somewhere he worked or still works for Lenovo.

If he was a Lenovo employee then all other devices he used as comparison would be Lenovo laptops.

 

[SiliconFly]

You forgot to mention that he retracted a lot of this stuff later, in the months leading up to MTL launch.

MLID publishing fake numbers, information and nonsense & later redacting false statements is just plain wrong. Not exactly what a decent professional would do. Misleading unsuspecting public is not the right thing to do in any and all contexts. And also, always claiming without any real reason or evidence that AMD is always superior this and that and Nvidia and Intel are useless, evil, bad and worthless is just brainless if you ask me.

I think Nvidia GPUs are better than AMD GPUs by a wide margin even today due to DLSS. They came out with ray-tracing 2 years before amd. They introduced framegen 1 year before amd. They have ray-reconstruction for which amd doesn't even have a solution yet. And claiming AMD is better at everything is just plain wrong. AMD has it's strength. But they have a long way to go before they can catch up with competition in both CPU & GPU imho.

 

[PJVol]

As actual information becomes available change your earlier predictions, which were of course how you got the views in the first place

And what's wrong about this?
The level of hatred towards him from a certain individuals here never ceases to amaze me, as if he eating someone else's lunch.

 

[SiliconFly]

Spreading misinformation and FUD is not good for anyone (except him of course).

I think the industry should come up with some effective mechanism to fight fake information being spread by bad actors.

 

[Abwx]

Spreading misinformation and FUD is not good for anyone (except him of course).

I think the industry should come up with some effective mechanism to fight fake information being spread by bad actors.

You mean like the firm that once talked of snake oil..?...

 

[SiliconFly]

You mean like the firm that once talked of snake oil..?...

According to MLID, if it fits in his picture... it's a fact. So, in his own logic, amd snake oil should be true. After all, Zen 2 cores are a fact!

Anyways, time to switch gears. Not worth wasting time on unproductive topics.

 

[maddie]

MLID publishing fake numbers, information and nonsense & later redacting false statements is just plain wrong. Not exactly what a decent professional would do. Misleading unsuspecting public is not the right thing to do in any and all contexts. And also, always claiming without any real reason or evidence that AMD is always superior this and that and Nvidia and Intel are useless, evil, bad and worthless is just brainless if you ask me.

I think Nvidia GPUs are better than AMD GPUs by a wide margin even today due to DLSS. They came out with ray-tracing 2 years before amd. They introduced framegen 1 year before amd. They have ray-reconstruction for which amd doesn't even have a solution yet. And claiming AMD is better at everything is just plain wrong. AMD has it's strength. But they have a long way to go before they can catch up with competition in both CPU & GPU imho.

Pot calling kettle?

 

[Hulk]

And what's wrong about this?
The level of hatred towards him from a certain individuals here never ceases to amaze me, as if he eating someone else's lunch.

I'm old fashioned, I believe in the old adage that goes like this...

Just because you have the right to do something doesn't mean it's the right thing to do.

That being said MLID is entertaining I will give you that. He also seems like a nice guy.

 

[SiliconFly]

Pot calling kettle?

Nope. AMD literally has a long way to go before catching up with competition. Intel has around 85% market share in client CPUs & Nvidia has around 90% market share in GPUs.

(I've applied your own font style. Bold & upsized with underline for clarity.)

 

[Hulk]

Uh ... regarding the Intel power specifications rabbit hole...

Since for Meteor Lake they don't specify base frequency what's the point of specifying base power?

Also can someone please explain "Minimum and Maximum Assured Power?"

I looked it up in Intel's docs and it basically says it's cTDP up/down. I think I'm missing something here?

Edit - Never mind, looks like they are just using new terminology as Max assured power is cTDP up now. Thing is, what's the point of providing these power specs if you don't provide the frequencies to which they refer?

Only turbo has meaning since frequency and power is provided. It almost seems like Intel is hiding something.

 

[Henry swagger]

According to MLID, if it fits in his picture... it's a fact. So, in his own logic, amd snake oil should be true. After all, Zen 2 cores are a fact!

Anyways, time to switch gears. Not worth wasting time on unproductive topics.

Mlid admitted to holding stock in amd... he bad talks nvidia and intel but owns a 13700k and 3070 lol

 

[SiliconFly]

Mlid admitted to holding stock in amd... he bad talks nvidia and intel but owns a 13700k and 3070 lol

There is one thing I find a bit shocking. A mod once said "criticize the post, not the poster".

But whats happening here is a bit different. When a forum member says something about MLID, I see a few members attack the poster instead of the post. Just trying to figure out why things happen the way its happening.

My 2 cents. Just an observation.

 

[Mopetar]

Anyone who hasn't figured out the MLID has a business mode that relies on throwing out a video related to any rumor, regardless of how unfounded it may be.

He doesn't care if he's wrong 100 times if he's right just once, and he probably doesn't care about being right once all that much either. As long as he gets clicks, he's getting what he wants.

He doesn't even care if people think he's an idiot. If people will hate watch his videos or link them just so they can dunk on him, he's getting more clicks.

He doesn't care if you troll him with bogus rumors that he reports on. He gets clicks.

 

[Meteor Late]

Geekerwan review:

Seems the main issue is the P core, E core looks decent, 10% more performance at 10% lower power, maybe would be roughly 15% higher performance at same power. P core however is garbage, those last 200-300MHz are very expensive on 12th and 13th gen, equalizing performance by power limiting / reducing the clock, power consumption would be similar based on what I've seen.
So yeah, MT efficiency disappointment is due to Redwood Cove being trash.

 

[Saylick]

Mlid admitted to holding stock in amd... he bad talks nvidia and intel but owns a 13700k and 3070 lol

Hey! That's only half true...




I think he has a 4090, actually.

 

[mikk]

If he was a Lenovo employee then all other devices he used as comparison would be Lenovo laptops.

His MTL review device is from Lenovo....

Geekerwan review:

Seems the main issue is the P core, E core looks decent, 10% more performance at 10% lower power, maybe would be roughly 15% higher performance at same power. P core however is garbage, those last 200-300MHz are very expensive on 12th and 13th gen, equalizing performance by power limiting / reducing the clock, power consumption would be similar based on what I've seen.
So yeah, MT efficiency disappointment is due to Redwood Cove being trash.

Battery life 32% above Raptor Lake battery normalized, interesting. Also interesting to note that the timespy graphics performance drops heavily down below 25W which is the same what happened in the golden pig test, something clearly doesn't work as it should.

Cinebench efficiency diagram same mediocre result as the golden pig test. Looks similar underwhelming to the pre updated pcode Cinebench results. If the pcode upgrade improvements can be confirmed it will look quite a bit different, same for the improved low power graphics driver update. It's too early to write MTL off.

 

[Meteor Late]

Not writing MTL off, mainly writing Redwood Cove off, P core is always the most important core, and it's just bad. Zen 5 will be an absolute bloodbath.

 

[TESKATLIPOKA]

Not writing MTL off, mainly writing Redwood Cove off, P core is always the most important core, and it's just bad. Zen 5 will be an absolute bloodbath.

Very likely yes, but let's be honest, It's kinda expected and wanted that the new product outperforms the older one, If It can't aka Bulldozer, then that's a big problem. MTL is not a bad product, personally expected more but whatever.
But now this is all old tech. I am already tuned for Strix vs ARL.

 

[FlameTail]

I am already tuned for Strix vs ARL.

Add the X Elite and make it a trio!

 

[FlameTail]

Geekerwan review:

Seems the main issue is the P core, E core looks decent, 10% more performance at 10% lower power, maybe would be roughly 15% higher performance at same power. P core however is garbage, those last 200-300MHz are very expensive on 12th and 13th gen, equalizing performance by power limiting / reducing the clock, power consumption would be similar based on what I've seen.
So yeah, MT efficiency disappointment is due to Redwood Cove being trash.

Geekerwan🔥