Discussion Intel current and future Lakes & Rapids thread

[jpiniero]

Well all you have to do is refer to history. Intel made a lot of strides in over 50 years, but they were outdone by AMD many times. Intel may be down for the next few years, but they'll come back eventually. Consumers should never favor one side over another or rather they shouldn't wish for Intel to be down forever or go away. We'll have another Core situation on our hands with AMD at the top. Competition is always good for the consumer.

Thing is, Intel's been able to rely on the fabs during the bad periods. Now... not so much.

You will still get competition but it might come from ARM and not Intel.

 

[lobz]

Thing is, Intel's been able to rely on the fabs during the bad periods. Now... not so much.

You will still get competition but it might come from ARM and not Intel.

But their CPU architecture teams are amazing right now.

 

[Exist50]

But their CPU architecture teams are amazing right now.

Nah. Single core performance is fine, but they're too big and power hungry.

 

[DrMrLordX]

Nah. Single core performance is fine, but they're too big and power hungry.

I don't think he's talking about anything on 14nm.

 

[Exist50]

I don't think he's talking about anything on 14nm.

I mean Sunny/Willow Cove as well.

 

[DrMrLordX]

I mean Sunny/Willow Cove as well.

Which Willow Cove product has proven itself to be powerhungry? We haven't even seen retail TigerLake yet.

 

[beginner99]

as I hate to plug several things

well docking implies a docking station or usb-c "dock" which means 1 thing to plugin.

At home I have a cheap travel usb-c dock and a usb switch so I can use the same mouse and keyboard for the work laptop or my personal desktop. pretty convenient.

 

[A///]

USB-C docks are cheap. If you're working with TB on an Intel or Apple product, then that raises the price considerably. I'll side with @lobz and say I also find it difficult to work on a laptop unless I can set it to an output monitor, keyboard and mouse.

 

[lobz]

well docking implies a docking station or usb-c "dock" which means 1 thing to plugin.

At home I have a cheap travel usb-c dock and a usb switch so I can use the same mouse and keyboard for the work laptop or my personal desktop. pretty convenient.

I probably forgot to mention that I also play Valorant semi-professionally on a 240 Hz display and there is NEVER enough fps when you play a shooter competitively, so having one laptop to act as all my computing needs everywhere is not an option

 

[A///]

The internet's favorite "engineer" is at it again, FWIW.

 

[eek2121]

TGL appears to be a great little product. Waiting to see power consumption, but the i7 performs better than a 4700U.

A lot of people are bashing Intel and making claims they can’t deliver. Clearly many of us have short memories. Time and time again Intel has been beaten, and they’ve always come back. Tiger Lake and Intel’s discrete graphics are the first indication that 10nm is beginning to ramp up. While 10nm is (extremely) late, we must keep in mind that until AMD makes the leap to 5nm, they won’t continue to have a process node advantage. With 5nm chips from AMD possibly not landing until 2H2022, Intel and AMD could very well find themselves on even footing from a process technology standpoint going forward, and believe it or not this is a good thing.

 

[DrMrLordX]

While 10nm is (extremely) late, we must keep in mind that until AMD makes the leap to 5nm, they won’t continue to have a process node advantage.

That's nice and all, but not only is it over three years late, it (being 10nm with "SuperFETs") still can't be found in anything except 4c mobile chips to come out in September. Rocket Lake-S isn't even out yet, and that's going to be 14nm again. The "true" 10nm - that is, something that might yield okay and not be hobbled by poor power characteristics beyond about 3.5 GHz - desktop and server products won't be showing up until 2021 (Alder Lake, Sapphire Rapids). Plus, Intel has so much confidence in the yield of 10nm + SuperFETs that they've ordered 180k 6nm wafers from TSMC.

I wouldn't pop the corks just yet.

 

[mikk]

It may be late but nobody expected such a big performance improvement from last year, most couldn't even believe it can go higher than 4.3 Ghz. I was hoping for 4.5 Ghz and no way I could believe it can go to 4.8 Ghz which is 14++ level of performance. Also RKL-U has been cancelled which means Intel is confident their Tigerlake 10nm volume is big enough by now.

 

[DrMrLordX]

RKL-U has been cancelled

Thank goodness. I can't see how that would have been a good product.

 

[jpiniero]

Also RKL-U has been cancelled which means Intel is confident their Tigerlake 10nm volume is big enough by now.

I'm sure volume of Tiger will be way higher than Ice, but you'd think that Rocket's demise just means that there will be more Comet.

 

[piokos]

That's nice and all, but not only is it over three years late, it (being 10nm with "SuperFETs") still can't be found in anything except 4c mobile chips to come out in September. Rocket Lake-S isn't even out yet, and that's going to be 14nm again. The "true" 10nm - that is, something that might yield okay and not be hobbled by poor power characteristics beyond about 3.5 GHz - desktop and server products won't be showing up until 2021 (Alder Lake, Sapphire Rapids). Plus, Intel has so much confidence in the yield of 10nm + SuperFETs that they've ordered 180k 6nm wafers from TSMC.

This post is all over the place.

People on this forum treat mobile chips like some poor byproduct of shiny desktop gaming chips. It's the other way round...
"4c mobile chips" is Intel's most important and profitable consumer lineup. If they're going to make it exclusively on Tiger Lake, it clearly shows a lot of confidence in both platform and process.

As for outsourcing production to TSMC: Intel just doesn't have enough production lines to make everything themselves.
It's absolutely a bummer and one of the reasons for worse forecasts and stock drop.
Still, outsourcing with lower their margins is better than handling OEMs to AMD.

If they kept making just CPU, their in-house manufacturing would probably manage. But they are expanding a few other niches lately: networking, car AI, FPGA and now GPUs.

 

[jpiniero]

"4c mobile chips" is Intel's most important and profitable consumer lineup. If they're going to make it exclusively on Tiger Lake, it clearly shows a lot of confidence in both platform and process.

But they aren't. There's going to be plenty of Comet Lake.

 

[mikk]

But they aren't. There's going to be plenty of Comet Lake.

Anything special about it? You can even buy Whiskey Lake, there are plenty of available. An old mobile generation won't disappear from one day to another, it will slowly disappear and this isn't Intel specific, there are more Picasso devices in the market than Renoir after like 6 months.

 

[jpiniero]

Anything special about it? You can even buy Whiskey Lake, there are plenty of available. An old mobile generation won't disappear from one day to another, it will slowly disappear and this isn't Intel specific, there are more Picasso devices in the market than Renoir after like 6 months.

I was thinking they would rebrand Comet Lake into 11th gen to sort of replace Rocket Lake-U, much like 10th gen. But they don't necessarily have to do that.

 

[itsmydamnation]

TGL appears to be a great little product. Waiting to see power consumption, but the i7 performs better than a 4700U.

What people are ignoring is that all the renior SKU's have focused on perf per watt over perf. What i mean is in 1T AVX stress test package power never goes beyond 11watts, 2T 19 watts, only at 4T does it go to 35watt.

I will be super surprised if intel is doing the same for TGL for single / low thread boost behaviour.

 

[IntelUser2000]

TGL appears to be a great little product. Waiting to see power consumption, but the i7 performs better than a 4700U.

We need to temper our expectations a tad.

Peak performance the 4700U will beat Tigerlake-U. There's just no way around having 8 cores. But it'll be pretty close.

I don't see it as coincidence reviewers receiving 4800U laptop samples just a week before Tigerlake launch.

I was thinking they would rebrand Comet Lake into 11th gen to sort of replace Rocket Lake-U, much like 10th gen. But they don't necessarily have to do that.

In what world do you think that makes sense? Maybe its a joke though. So this is my response.

 

[jpiniero]

In what world do you think that makes sense?

Same reason they did it with 10th gen - need the volume and OEMs hate selling 'last gen' products.

 

[IntelUser2000]

Same reason they did it with 10th gen - need the volume and OEMs hate selling 'last gen' products.

They did it with 10th Gen because Cometlake had a reason to exist - 6 cores.

Cometlake-U has no reason to exist this generation.

-Value: Pentium and Celerons are coming. Not on Icelake
-vPro: That's next year, also not on Icelake
-IoT Tigerlake leaks. Nope, not on Icelake

 

[LightningZ71]

It also still remains to be seen how Tiger Lake's Xe graphics will perform in real benchmarks. Unless it's a complete blowout, there will be room for AMD to release 4550, 4650, and 4750u SKUs that have one additional CU enabled as opposed to the original versio to allow them to close or tighten the gap as needed. While I do expect Xe to be quite good, it's got a lot of ground to make up against AMD.

I don't see it as impossible that the top end i7 Tiger Lake could boost significantly higher than the 4700u for short periods. I don't see it as impossible that, while in that boost phase, the Four core i7 Tiger Lake could conceivably turn in higher multicore performance numbers than the 4700u. I would be shocked if the TGL i7 can sustain higher multicore performance over long durations or sustain higher gaming performance than the 4700u over longer durations.

 

[vstar]

Intel Core Gen11 TigerLake ULV (i7-1165G7) Review & Benchmarks – CPU AVX512 Performance – SiSoftware

www.sisoftware.co.uk

Surprised to see this review early, but TGL looks like a beast!

Quoting from the article in case it is taken down:

"While due to high power consumption, AVX512 was never a good fit for mobile and their meager power envelopes (15W, 25W, etc.) – somehow “Tiger Lake” (TGL) manages to run them much faster, 40-50% faster than “Ice Lake” and thus beating the competition.

TGL’s performance with a meager 15W power budget in a think & light laptop is pretty compelling and it soundly beats not only older (bigger) mobile processors with more cores (at 35-45W) but also older desktop processors! It is truly astonishing what AVX512 can bring on a modern efficient design."