Discussion Intel current and future Lakes & Rapids thread

[uzzi38]

How exactly is "presentation from an HPC conference" a leak?

Seriously, AMD has been called out on hyping (even very distant) products for many years. I don't understand your motivations, but neglecting that just makes you look silly. Don't go that path.

Just remind yourself how the year before Zen looked. And with Zen2 launch AMD said they have 2 leapfrogging design teams - basically saying that they'll be hyping 2 future generations instead of one.
And what about RDNA2? How long has that been going on?

As @dullard said: hyping products is part of this business. And AMD has been doing this more simply because they were building the image of an enthusiast brand.
In reality neither of these brands really focuses on the DIY enthusiast niche. It's just that Intel doesn't even give a f... to fake it (something I appreciate, to be honest).

Because it was never supposed to be a client facing presentation? It was supposed to be with strict HPC customers only?

We - the public - should have never had access to that presentation. There's a reason it was taken down in hours.

 

[mikk]

Back in the old days Intel never released a new generation within less than 12-15 months but since AMD has the lead in desktop (beside gaming) and is coming with Zen 3 very soon they cannot wait and see their market share drop every week and month. On Steam they are losing -0.5-1.0% each month to AMD. They have a bit more time in mobile since TGL-H is coming with 8C. ADL desktop is clearly their lead platform. Which is suspect also....the first Rocket Lake ES in the wild was spotted in April, a small first ADL-S ES entry in July and a bigger one in October this month. This is suspiciously close.

 

[arandomguy]

My point is that teasing and generating hype for products 6 months ahead of when they will actually launch us a trend that should die. And I said wait a week or two so you can see that in action

I've always been curious about this mind set and am wondering if those that feel this way can explain why they do so?

At least for me personally I want as much information as far out as possible. If Intel or AMD (or whomever) want to detail out their plans for a product 5 years from now I'd love it, even if it will obviously experience changes when it finally comes out. If AMD wants to expouse on details on what Zen 4 will have over Zen 3 before Zen 3 is even avaiable that's great in my opinion. At the very least I don't see how it hurts the consumer on a practical level.

 

[piokos]

Because it was never supposed to be a client facing presentation? It was supposed to be with strict HPC customers only?

We - the public - should have never had access to that presentation. There's a reason it was taken down in hours.

Why not?
Maybe you aren't interested in HPC/datacenter stuff but someone else is. It was an official presentation shown to the public.
HPC companies/institutions are AMD's clients just like you are.

This could have been called "a leak" if the slides came from a particular HPC company or OEM, i.e. this would be an internal, confidential information.
A conference is a conference. Everyone can go. They don't ask you to turn your phone off, nor forbid you to tell anyone what you've seen.

 

[uzzi38]

Why not?
Maybe you aren't interested in HPC/datacenter stuff but someone else is. It was an official presentation shown to the public.
HPC companies/institutions are AMD's clients just like you are.

This could have been called "a leak" if the slides came from a particular HPC company or OEM, i.e. this would be an internal, confidential information.
A conference is a conference. Everyone can go. They don't ask you to turn your phone off, nor forbid you to tell anyone what you've seen.

Wha-

No, it was a video that was supposed to be unlisted. They completely took it down when they realised it wasn't.

EDIT: Or maybe private, idk the differing levels of YT preferences

 

[piokos]

Wha-

No, it was a video that was supposed to be unlisted. They completely took it down when they realised it wasn't.

EDIT: Or maybe private, idk the differing levels of YT preferences

So was it something that shouldn't be shown during the conference?

Because, once again, showing something on a conference means it's an announcement, not a leak.
Of course a company is entitled to remove some materials from the Internet (as much as they can). But if you show something and change your mind afterwards, it's still an announcement.
A leak happens when confidential information is released without authorization.
We're pretty liberal with this anyway, given that some "leaks" are orchestrated by the companies.

Anyway, I still don't understand why anyone would think early announcements are bad for the market?
How on Earth can extra pre-launch information be a problem?

Intel shares a lot of information with OEMs way before anything becomes public. This is most likely true for Nvidia as well, given how quickly graphics cards become available.
Some (most) of this leaks before the official launch. Intel has to accept this.

AMD often keeps some details for the launch, so the leaks naturally can be less common.
But seeing that someone would call that better (for whatever reason) is really surprising...

 

[mikk]

Asus Zenbook S 13.9 inch https://www.ultrabookreview.com/wp-content/uploads/2020/09/back-3.jpg
Acer Swift 3 13.5 inch https://www.ultrabookreview.com/wp-content/uploads/2020/10/back.jpg


No wonder the Acer has a much better cooling, the air intakes are tiny in comparison with Acer. Not a good one from Asus, I would expect something better from a premium model.

 

[coercitiv]

No wonder the Acer has a much better cooling, the air intakes are tiny in comparison with Acer. Not a good one from Asus, I would expect something better from a premium model.

It seems someone at Asus does have a habit of understimating the importance of good air intake. They were recently criticized by Hardware Unboxed for the rather limited intakes on one of their TUF gaming laptops, and I can also confirm they used to keep intakes to minimum as far as Haswell laptops. Not only did my Asus unit have limited intake grills, but they were further reduced by a second layer of plastic mesh which probably had esthetic purpose.

 

[mikk]

It's a pity because otherwise they have a big range of Tigerlake models and I can't buy a device with such a poor cooling. This is good enough for 15W but surely not for 20W+
Acer isn't premium and has better cooling, Asus should do better.

 

[DrMrLordX]

If true, may as well just kill it and try to release Sapphire Rapids in late 2021.

Agreed. Though it's hard to kill what was never alive.

Anyone else feel that Alder Lake is completely wrong product, that bets performance on Windows process scheduler with barely anything to show for in return?

Not necessarily.

At it's base, Alder Lake looks like a 2c-8c CPU that has some Gracemont thrown in to imitate the ARM/Android world. There are ARM SoCs out there with bizarre 10c configurations and multiple core types, and Android doesn't seem to have many problems figuring out what to do with which cores. The Gracemont cores are really just there for power-saving and picking up low-priority background threads.

Thinking of Alder Lake as anything other than a (at most) 8c/16t CPU is really making things too complicated.

 

[piokos]

It's a pity because otherwise they have a big range of Tigerlake models and I can't buy a device with such a poor cooling. This is good enough for 15W but surely not for 20W+
Acer isn't premium and has better cooling, Asus should do better.

Back when laptops were fat, putting vents on the bottom was seen as a big flaw. Everyone knew this.
I don't know if you forgot about it or you don't use a laptop on the go...

It's slightly harder to design the slim laptops we have today, but it doesn't mean we should become lenient.

Acer's design is actually quite bad. When put on a flat surface, the bottom outflow will be blocked by the massive back stand.
It doesn't matter that the vent itself is large - the air still has to go through a slim gap at the sides - which is another problem, because it'll means hot air is pushed onto your hands (especially when using a mouse).
Also, some of the hot air will go directly into the intakes.
When put on laps, a bed or any fluffy surface, it'll quickly choke.

Asus did this much better: small intakes near the front and outtake at the back. Minimal impact from what you put it on and hot air exiting where it's least likely to go back at you or into the intakes. Very nice design.

You may think Acer's design is similar to the universally praised Dell XPS 13.
However, the bottom vent on the XPS is an intake. Outtake is in the back panel (just like on Zenbook).





This is how it should be done.

 

[mikk]

Back when laptops were fat, putting vents on the bottom was seen as a big flaw. Everyone knew this.
I don't know if you forgot about it or you don't use a laptop on the go...

It's slightly harder to design the slim laptops we have today, but it doesn't mean we should become lenient.

AMDs showcase device Yoga Slim 7: https://www.ultrabookreview.com/wp-content/uploads/2020/09/back-2.jpg

Nobody complained...

Asus did this much better: small intakes near the front and outtake at the back. Minimal impact from what you put it on and hot air exiting where it's least likely to go back at you or into the intakes. Very nice design.

The Asus cooling is by far the worst at the moment, higher fan noise and higher temperatures inside+outside, check the review and repeat it's a good cooling.

I was expecting something else, given the thermal module in the older UX392 series, and I’m actually surprised this performs the way it does with this kind of cooling, especially since Asus also went for a close back and limited air-intake on the bottom panel, restricted to those narrow grills on the sides. Air is also sucked in through the keyboard and from behind the hinge, though.

As for the exhaust, it blows the hot air into the screen, and while some of it is absorbed by the display’s hinge, the panel itself reaches temperatures of 50+ degrees C in the area around the exhaust, something I’m not comfortable with long-term. If you’re planning on running demanding loads and games on this laptop, I’d carefully consider this aspect in my purchase decision.

Sounds really good

 

[IntelUser2000]

I think its more than thermals that make the Asus suck. There are finer details that we don't know. Like how the firmware handles boost and thermals. Manufacturers have more than a dozen knobs they can adjust.

 

[coercitiv]

Acer's design is actually quite bad. When put on a flat surface, the bottom outflow will be blocked by the massive back stand.
It doesn't matter that the vent itself is large - the air still has to go through a slim gap at the sides - which is another problem, because it'll means hot air is pushed onto your hands (especially when using a mouse).
Also, some of the hot air will go directly into the intakes.
When put on laps, a bed or any fluffy surface, it'll quickly choke.

You may think Acer's design is similar to the universally praised Dell XPS 13.
However, the bottom vent on the XPS is an intake. Outtake is in the back panel (just like on Zenbook).

What on earth are you talking about about?!

The Acer sucks air from the bottom and pushes air through the back towards the top. It's definitely not the best cooling design in the industry, but you somehow managed to imagine it works in reverse.

 

[IntelUser2000]

Post-Golden Cove is like Willow Cove and supposed to have minimal gains.

AMD and ARM are getting 15-20% per year.

Intel seems extremely conservative in this regard. Golden Cove needs 30-40%, not 20% to keep in line.

 

[inf64]

Post-Golden Cove is like Willow Cove and supposed to have minimal gains.

AMD and ARM are getting 15-20% per year.

Intel seems extremely conservative in this regard. Golden Cove needs 30-40%, not 20% to keep in line.

There was a leak on twitter supposedly projecting Golden Cove to have similar jump as Icelake (~18-20%). That would put them somewhat ahead of Zen3 but Zen4 is coming very soon after or maybe around the same time, on 5nm with 2x perf./watt versus Zen3. It's an uphill battle for intel, they need their fabs to deliver.

 

[mikk]

Acer Swift 3 14 i5-1135G7: https://www.ultrabookreview.com/42008-acer-swift-3-14-sf314-59-review/

i5-1135G7 beats i7-1165G7 in gaming:

As suggested by the benchmarks results above, this Swift 3 14-inch ends-up outmatching the other i7-1165G7 models we’ve tested so far in real-life gaming as well. Power allocation makes a big difference here and allows the iGPU to run at its peak frequencies of 1.3 GHz in most of the titles, with fairly solid CPU frequencies as well.

80 Xe LP EUs are enough to easily beat AMDs fastest 4800U device Yoga Slim 7 if it's a decent 28W device.

 

[IntelUser2000]

There was a leak on twitter supposedly projecting Golden Cove to have similar jump as Icelake (~18-20%). That would put them somewhat ahead of Zen3 but Zen4 is coming very soon after or maybe around the same time, on 5nm with 2x perf./watt versus Zen3. It's an uphill battle for intel, they need their fabs to deliver.

They are definitely on the losing side.

Alderlake should have been a 4 Golden Cove, 16 Gracemont part to maximize the idea of hybrid scheme. And those Golden Cove cores should have been extra fat, for Core+.

 

[inf64]

They are definitely on the losing side.

Alderlake should have been a 4 Golden Cove, 16 Gracemont part to maximize the idea of hybrid scheme. And those Golden Cove cores should have been extra fat, for Core+.

I agree, 8+8 will lose (probably badly) in MT workloads against Zen3 16/32T parts. I still expect Golden Cove to be impressive core, still it could be too little too late as AMD is really fired up and continuously delivered without one hitch since 2017, which is impressive.

 

[mikk]

There was a leak on twitter supposedly projecting Golden Cove to have similar jump as Icelake (~18-20%). That would put them somewhat ahead of Zen3 but Zen4 is coming very soon after or maybe around the same time, on 5nm with 2x perf./watt versus Zen3. It's an uphill battle for intel, they need their fabs to deliver.

The projection is a projection, this is not based on some real infos from Intel. And by the way next up is Warhol after Vermeer which is Zen3 and 7nm based, means Warhol will go against Alder Lake-S. Hower RKL-S seems lost overall, no wonder Intel pushes Alder Lake-S and not Alder Lake-P.

 

[jpiniero]

I agree, 8+8 will lose (probably badly) in MT workloads against Zen3 16/32T parts. I still expect Golden Cove to be impressive core, still it could be too little too late as AMD is really fired up and continuously delivered without one hitch since 2017, which is impressive.

Which is fine since Intel doesn't sell mainstream parts at the 16C price point. That's more for HEDT, the real competitor would be a theoretical Sapphire Rapids-X.

 

[IntelUser2000]

I agree, 8+8 will lose (probably badly) in MT workloads against Zen3 16/32T parts. I still expect Golden Cove to be impressive core, still it could be too little too late as AMD is really fired up and continuously delivered without one hitch since 2017, which is impressive.

Actually, I expect if the two setups can work together it'll be far more competitive than Cometlake was against Zen 2.

 

[IntelUser2000]

80 Xe LP EUs are enough to easily beat AMDs fastest 4800U device Yoga Slim 7 if it's a decent 28W device.

Tigerlake changes the way PL1 works.

Before, 28W PL1 means the application can use up to 28W long term, and the more demanding the application, the less it can stay short term before it has to go down to the PL1 load.

On Tigerlake, it could be 28W PL1 on Cinebench, but 25W on Prime 95, and maybe even lower on LinX.

This variable PL1 scheme seems to have one advantage at least. Previously when gaming, you'll get good frames(say 40 fps), but few mins later it'll go down to 30 fps and become choppy.

Variable PL1 means it'll lock onto one TDP limit(thus frames per second) and stay there.

 

[inf64]

Actually, I expect if the two setups can work together it'll be far more competitive than Cometlake was against Zen 2.

Oh there is no doubt about that, Comet Lake is long in the tooth, I bet intel can't wait to replace it, even with Rocketlake. I hope all of this renewed competition will give us much better price/perf parts especially in the laptop space. If anyone told me 2 years ago that we would have 8C laptops with such performance as Renoir or such a high ST performance as Tigerlake has, I would have dismissed it easily.

 

[mikk]

Asus Zenbook S 13.9 inch https://www.ultrabookreview.com/wp-content/uploads/2020/09/back-3.jpg
Acer Swift 3 13.5 inch https://www.ultrabookreview.com/wp-content/uploads/2020/10/back.jpg


No wonder the Acer has a much better cooling, the air intakes are tiny in comparison with Acer. Not a good one from Asus, I would expect something better from a premium model.

Three more...

Acer Swift 5 14 inch i7-1165G7: https://geex.x-kom.pl/wp-content/uploads/2020/09/acer-swift-5-spod-urzadzenia.jpg
Acer Swift 3 14 inch i5-1135G7: https://www.ultrabookreview.com/wp-content/uploads/2020/10/back-1.jpg

MSI Prestige 14 Evo reference ES: https://images.idgesg.net/images/article/2020/09/inkedpxl_20200915_203905705_li-100857614-orig.jpg

The Intel Tiger Lake reference laptop we used featured two heat pipes and a single fan and fin stack.

Intel 11th-gen Core i7-1185G7 Tiger Lake Preview: It’s (mostly) faster than Ryzen

11th-gen Core i7-1186G7 vs. Ryzen 7 4800U: Who wins?

www.pcworld.com

The MSI Prestige (ES) device in the reference test did have big air intakes, however it didn't use a two fan solution unlike the Yoga Slim.