Review 'The Microsoft Surface Laptop 3 Showdown: Amd Picasso vs Intel Ice Lake' - Anandtech

[Gideon]

Another example that Picasso inherently does not have bad battery life, it's more down to the laptop implementation:

Lenovo IdeaPad S540-13API Laptop in Review: Slim Ryzen Companion with Long Battery Life

We are testing the Lenovo IdeaPad S540, configured with an AMD Ryzen 5 3550H, AMD Radeon RX Vega 8, 16 GB of RAM and a 512 GB SSD.

www.notebookcheck.net

Battery Life
In our real-world Wi-Fi test, the IdeaPad S540-13API lasts around ten and a half hours. This is an above average result compared to the competitors we chose and hints at an effective use of the 65-Wh battery. At maximum brightness, it still reaches a runtime of just under seven hours.
...
In our Wi-Fi test, the mobile laptop shines with a very long battery life and the input devices feel comfortable to use as well.

 

[lobz]

Another example that Picasso inherently does not have bad battery life, it's more down to the laptop implementation:

Lenovo IdeaPad S540-13API Laptop in Review: Slim Ryzen Companion with Long Battery Life

We are testing the Lenovo IdeaPad S540, configured with an AMD Ryzen 5 3550H, AMD Radeon RX Vega 8, 16 GB of RAM and a 512 GB SSD.

www.notebookcheck.net

You may piss off some members here with this 'opinion'.

 

[RetroZombie]

Thank you for the link gideon.

Here are some interesting comparison graphs:


AMD Ryzen 5 3550H, high tdp cpu 35 Watts (notebook check)



Intel Core i5-1035G4, low tdp cpu 15 Watts (notebook check)


Even an amd 35W cpu have a much lower load power consumption vs intel 15W cpu, now imagine if the amd system could consume the same amount of power wouldn't the performance gap decrease and intel no longer would look so good right?

 

[LikeLinus]

Another example that Picasso inherently does not have bad battery life, it's more down to the laptop implementation:

Lenovo IdeaPad S540-13API Laptop in Review: Slim Ryzen Companion with Long Battery Life

We are testing the Lenovo IdeaPad S540, configured with an AMD Ryzen 5 3550H, AMD Radeon RX Vega 8, 16 GB of RAM and a 512 GB SSD.

www.notebookcheck.net

AMD Laptop = 65 Wh battery
Intel Surface Laptop 3 = 45 Wh Battery

AMD last 10.5 hrs with a 65wh battery in that review.

Intel lasted 10.2hrs with a 45wh battery, but also it is an i7 and not an i5. Much higher performance and just as long battery life, but with a smaller battery. Not even getting into the fact that the Surface has a large screen and so on.

No one is saying it has inherently bad life, but it's not nearly as power efficient as the Intel based laptop.

What all this has to do with the OP and discussion, who know. The thread is about the Surface Laptop 3.

 

[IntelUser2000]

No one is saying it has inherently bad life, but it's not nearly as power efficient as the Intel based laptop.

What all this has to do with the OP and discussion, who know. The thread is about the Surface Laptop 3.

It's not a good idea to compare results from different sites. They use their own testing methods and will work out differently.

Also, because implementations can vary, you have to take the average, as some will be worse than others. Yes, its difficult, because there are dozens of models out there.

 

[LikeLinus]

It's not a good idea to compare results from different sites. They use their own testing methods and will work out differently.

Also, because implementations can vary, you have to take the average, as some will be worse than others. Yes, its difficult, because there are dozens of models out there.

Yeah, no kidding. You didn't notice the multiple post showing different laptops with completely different configurations?

But the Anandtech (and others) article is from the exact same laptop with the same configuration between the top of the line Intel and AMD cpus. Why these guys are throwing around all this other junk, which is completely off-topic, I have no clue.

 

[RetroZombie]

Why these guys are throwing around all this other junk, which is completely off-topic, I have no clue.

Maybe you should read all the thread, since the beginning because I'm on topic and the others providing links to others reviews are also too.

A couple of points:
LikeLinus incorrectly stated battery capacity and forgot to mention battery run time:

AMD Laptop	56 Wh battery	627 minutes
Intel Surface Laptop 3	45 Wh Battery	490 minutes

Posting wrong information is not correct.

The thread is about the surface laptop and it's specific components, I can't compare it with any other, why? If notebook check for example would review 1000 microsoft surface 3, all the exact same models only then I would be on topic?

It's you who said:
But the Anandtech (and others) article is from the exact same laptop with the same configuration between the top of the line Intel and AMD cpus.

Are you sure about that, all the exact same, AMD does the exact same of intel? What do they share in common, the keyboard?

 

[LikeLinus]

Maybe you should read all the thread, since the beginning because I'm on topic and the others providing links to others reviews are also too.

A couple of points:
LikeLinus incorrectly stated battery capacity and forgot to mention battery run time:

AMD Laptop	56 Wh battery	627 minutes
Intel Surface Laptop 3	45 Wh Battery	490 minutes

Posting wrong information is not correct.

I quoted Gideon's article.

"Battery Life
In our real-world Wi-Fi test, the IdeaPad S540-13API lasts around ten and a half hours. This is an above average result compared to the competitors we chose and hints at an effective use of the 65-Wh battery. At maximum brightness, it still reaches a runtime of just under seven hours"

The article reads as they are suggesting that it effectively uses a 65wh battery.

Additionally, the Intel Surface Laptop 3 runs for 614 mins, not 490. The AMD version runs at 450mins. That's what I posted from the Anandtech article.

This IS battery run life doing web surfing at 200nits (brightness of display). And completely on-topic because it's the two laptops in the OP and article. Intel runs for almost 2 hours longer.

But thanks for making my point. The AMD laptop is using a 65wh batter with a slower CPU and the Surface is using a more power CPU and smaller battery and their battery life is almost the same.

Surface is a better base line because they use the same battery, screen, hard drive, audio, trackpad and various other internals. No test bed is going to be completely 100% the same, but you have someone posting completely different vendor laptop results trying to draw conclusions. At least using the same vendor and laptop model makes it an even playing field. We won't even get into the fact that the TDP you are mentioning, that is the recommended TDP. The actual TDP of that AMD chip and the Intel chips can go down to 12tdp. They can be adjusted by the manufacture and you do not know what they are actually running at. That can greatly affect battery life.

Spin this however you want and make up all the excuses you want. The whole "it's not the exact same laptop!". It's basically scraping the bottom of the barrel to find any excuse to make your argument. It's all a bit silly. But, continue on, if you must.

 

[IntelUser2000]

@LikeLinus You have made some good points.

NBC did make an error though and switched the 6 and 5. It's indeed 56WHr, but overall you are right.

AMD Ryzen 5 3550H, high tdp cpu 35 Watts (notebook check)

Intel Core i5-1035G4, low tdp cpu 15 Watts (notebook check)

Even an amd 35W cpu have a much lower load power consumption vs intel 15W cpu, now imagine if the amd system could consume the same amount of power wouldn't the performance gap decrease and intel no longer would look so good right?

Their Surface Laptop 3 with a "15W" 3780U uses 7-8W more power. Also, the 1035G4 according to stress tests are set at 22W.

The CPU does not throttle on battery for the SL3, while it does for the Lenovo 3550H system.

 

[IEC]

I imagine Microsoft is using the Zen+ based Picasso to work out any issues and familiarize themselves with optimization for AMD before Zen 2 based APUs launch in 2020. So while Picasso is decidedly and objectively inferior to Ice Lake in this application, I suspect Ice Lake is going to get its comeuppance soon enough.

 

[RetroZombie]

@LikeLinus NBC did make an error though and switched the 6 and 5. It's indeed 56WHr, but overall you are right.

Which immediately makes all he said wrong...

I quoted Gideon's article.

You quote I will take print screens:

Or we are both wrong, only you are or It's me.

Surface is a better base line because they use the same battery, screen, hard drive, audio, trackpad and various other internals.

The only 'similar' thing it's the brand, the os, ...

No test bed is going to be completely 100% the same, but you have someone posting completely different vendor laptop results trying to draw conclusions.

That's how you buy something you compare, that's why there are tests other wise they are all pointless. You compare cpu A with TDP of 65W but consumes 1000W vs cpu B with the same 65W TDP but it uses 65W, oh wait I can't do that, it's incorrect to compare A to B according to you.

At least using the same vendor and laptop model makes it an even playing field. We won't even get into the fact that the TDP you are mentioning, that is the recommended TDP. The actual TDP of that AMD chip and the Intel chips can go down to 12tdp. They can be adjusted by the manufacture and you do not know what they are actually running at.

According to intel it does not matter, I posted a link above if you had read it you weren't saying this. Even if the laptop vendor selected the 12 W tdp, because of PL1 setting it's allowed to use 45 Watts.

I don't have a problem to reply to your excuses, don't you worry.

 

[LikeLinus]

You quote I will take print screens:
View attachment 14995

You're comparing a laptop with a larger batter and Ryzen 5, not the Ryzen 7, against the i7 with a smaller battery. Not to even mention those results don't seen right. You do realize that 56wh is a larger battery than the 45wh batter, right? So the expectation that it would have longer battery life, wouldn't be surprising. But we won't even get into the fact that the Surface runs a touch screen, higher resolution, higher performance CPU, better Wifi and so on.

Take a look at the battery life graphic I posted from Anandtech's review of the Surface Laptop 3. If you choose to think it's not real, doesn't matter, doesn't fit your intended results, I honestly don't care at this point.

"The only 'similar' thing it's the brand, the os, ..."

OK buddy, I'm done replying to all of this non-sense. It's completely off-topic. Enjoy your evening.

 

[LikeLinus]

I imagine Microsoft is using the Zen+ based Picasso to work out any issues and familiarize themselves with optimization for AMD before Zen 2 based APUs launch in 2020. So while Picasso is decidedly and objectively inferior to Ice Lake in this application, I suspect Ice Lake is going to get its comeuppance soon enough.

No doubt. Microsoft is waiting for AMD to get their mobile CPUs up-to-date. But for some reason RetroZombie is unable to understand the fact that the current line of CPUs are inferior. Hopefully they will catch up early-mid next year with new mobile chips.

 

[krumme]

No doubt. Microsoft is waiting for AMD to get their mobile CPUs up-to-date. But for some reason RetroZombie is unable to understand the fact that the current line of CPUs are inferior. Hopefully they will catch up early-mid next year with new mobile chips.

The comparison is wrong as its products that adresses completely different markets. The idea that the difference is 100 usd is nonsense as can be seen in laptops actually sold in numbers.

In my house we have a Lenovo thinkpad t460 skl i5 and t480 (quad 8250) and a thinkpad 495 with the 3500u. If there is any battery life difference its certainly down to the aprox 1w difference as notebookcheck meassures and i cant subjectively tell any difference. And the power difference is only noted in games where the Vega is sometimes twice as fast. The rest is more or less the same.

Picasso 3500u is in my experience an extremely balanced apu for a low or midrange consumer laptop - not an expensive business machine. There will come some fance new expensive stuff, but this dirt cheap apu is a damn miracle for the low and mainstream comsumer laptops. Twice the cpu and gpu power as Intel laptops just 2 years ago, and at low price - The thinkpad 495 was literally half price the t models. And it has a battery life that is perfectly fine for most consumers. It will sell for years.

 

[clemsyn]

My Dell 3493 with an i5-1035G4 beats the Ryzen 7 in Cinebnech R20 Multithread

 

[DrMrLordX]

It will sell for years.

I hope not. AMD should have their Picasso-based products replaced by Renoir pretty soon. No need to keep selling 12nm stuff.

 

[mikk]

No doubt. Microsoft is waiting for AMD to get their mobile CPUs up-to-date. But for some reason RetroZombie is unable to understand the fact that the current line of CPUs are inferior. Hopefully they will catch up early-mid next year with new mobile chips.

I think everyone with some knowledge is aware that AMDs power management in low load load or idle isn't as good as Intels in the mobile space. AMD made improvements with Picasso and some of the implementations are really good nowadays but overall Intel is superior in this regards.

 

[Arkaign]

The showstopper for this whole thing is we've already seen what true 7nm Zen2 means vs Zen+ derivatives : a monumental improvement that pushed AMD from a decent alternative to the defacto best product outside of edge scenarios.

This old stuff sitting around carrying '3000' clothes in the laptop segment is just kind of wasting our time while we wait for the real offerings to arrive. The weird thing will be model numbers. Will they give it '4000' designation despite actual Zen3 not arriving for many months afterwards? I am beginning to really dislike this era of product naming.

 

[krumme]

I hope not. AMD should have their Picasso-based products replaced by Renoir pretty soon. No need to keep selling 12nm stuff.

You prefer Britol ridge to Picasso?
They will co exist for different segments. Amd have to move a lot of gf capacity due to wsa, and picasso is perfectly fine for the low end. It will eventually pressure out Bristol ridge. My bet is picasso will stay longer than next gen zen2 apus and even coexist wit zen3 and zen 4 apus. Wsa, dirt cheap ddr4.

 

[moinmoin]

The showstopper for this whole thing is we've already seen what true 7nm Zen2 means vs Zen+ derivatives : a monumental improvement that pushed AMD from a decent alternative to the defacto best product outside of edge scenarios.

I agree with the model naming stuff being inane. But the above (APUs being last gen where the current gen rises expectation in the upcoming APUs with current gen cores) will be true regardless of the naming. APUs will always come after the respective CPU gen launches. CPUs can be picked up by the DIY market right away whereas APUs depend on OEMs to have decent TTM so will quickly be "old hats" as soon as the next Zen gen is already closing in. This is the "downside" of AMD's current quick cadence.

 

[Tup3x]

E495 uses only single channel memory. Remove other dimm for Surface Laptop and idle power should be similar.

For E495 2x8GB is option too. (Though in that test it was single stick variant.)

 

[DrMrLordX]

You prefer Britol ridge to Picasso?

No, I prefer Renoir to Picasso. If AMD has to keep taking wafers from GF (which I doubt), then of course they can use Picasso to unseat Bristol Ridge. But Renoir should be the product in the Microsoft Surface 3+ or Surface 4 or whatever it is that MS releases next. Not Picasso.

My expectation is that AMD will move to Renoir/Dali and cover their entire lineup with that. Van Gogh for a very specialized market segment. Picasso, Bristol Ridge, and Stoney Ridge should all be tossed at once (and my expectation is that Bristol/Stoney are already gone).

 

[RetroZombie]

Picasso, Bristol Ridge, and Stoney Ridge should all be tossed at once (and my expectation is that Bristol/Stoney are already gone).

I would hope they would keep them around to fight the intel atom junk.
I was using one of my friend intel N2840 from 2014 to do some maintenance and was surprised how still usable is since it has an HDD on it, and another friend with a new N3000 (or N4000 cant remember) and was a total unusable junk and it's from 2019...

And since the athlon 300u already made all the intels 7th gen i3 to i7 look obsolete any way.

The amd A6/A9 and up are more than usable, still fast enough, the A4 an E2 however felt similar to the atom N3000/N4000, slightly faster but sill unusable, it's these types of products that amd needs to get rid off, they leave you a 'bad taste' in your mouth damaging your brand name reputation.

The bulldozzer APU's always felt like low end when used in desktop cpus but in mobile they where always a level up, many models could keep up with the i5 and sometimes near i7.

 

[RetroZombie]

But for some reason RetroZombie is unable to understand the fact that the current line of CPUs are inferior.

All the time I mentioned TDP, power consumption, ... not performance. But they some how are related:

CPU​	TDP​	Power consuption peak​
I9-9900K​	95W​	210W​
I9-10900K​	125W​	250W​
I7-1065G7​	15W​	45W​
i5-1035G4​	15W​	45W​
R9-3950X​	105W​	?​
R5-3550H​	35W​	35W​
R7-3780U​	15W​	?​

Feel free to say what is the value at ?
AMD with renoir could say it's 15W TDP and feed the 8 core monster with 45W and I want to see if you will complain or not.

 

[DrMrLordX]

I would hope they would keep them around to fight the intel atom junk.

I thought that's what Dali was for? And maybe the low-end Renoir products?