• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Discussion Intel current and future Lakes & Rapids thread

Page 280 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

ondma

Golden Member
Mar 18, 2018
1,572
447
106
Brutal ST performance as expected from GB5 leaks. One has to wonder how well it would score with desktop class memory setup.
Yea, single thread performance is outstanding. Too bad they dont have more cores in mobile (I know it is supposed to be coming), and the same sort of IPC for the desktop.
 

mikk

Platinum Member
May 15, 2012
2,912
730
136
Interesting that Tigerlake and Renoir when both limited to 15W perform about the same with Renoir seeming to have the edge
Which tests? At pcworld Tigerlake PL1 15W win 3 of 4 games against 4800U PL1 38W.



It had LPDDR4X 4266, that's pretty good.

It isn't because the latency is extremely poor for desktop standards.
 

Markfw

CPU Moderator, VC&G Moderator, Elite Member
Super Moderator
May 16, 2002
20,812
8,997
136
In the Toms review, it seems the only times the Tigerlake won, was at 28 watt, all the 15 watt (like the 4800u) it lost.
 
  • Like
Reactions: Drazick and Tlh97

JoeRambo

Senior member
Jun 13, 2013
873
569
136
It had LPDDR4X 4266, that's pretty good.
Not really. Desktop class memory has like 50ns latency even in useless "stock" tests like Anandtech does. 90ns for TGL is really nice for mobile, but twice as slow as proper desktop class memory ( talking about DDR4 3800CL16 class stuff here, not CL30+ ).
 
Last edited:

Hitman928

Platinum Member
Apr 15, 2012
2,914
2,453
136
Tom's tested 5 games and 15W Renoir won in all 5. Anandtech tested 5 games and Renoir won in 3/5. NBC tested only 2 games with both at 15W, Tigerlake wins 1 by 10% but Renoir wins the other by 38%. Like I said, they're close but from what I've seen, Renoir has the advantage at 15W. That really wasn't the point of my post though, the point was that there's a huge discrepancy in performance going from 15W Tigerlake iGPU to 28W Tigerlake iGPU compare to Renoir moving up in power.

It will be interesting to see how TGL does in OEM machines as well. Hopefully we get a like for like comparison between the two in something like the Yoga slim that many reviewers used for Renoir. Will TGL performance change once OEMs try to adjust for more battery life? What will TGL battery life look like? Still lots of info to come.
 

naukkis

Senior member
Jun 5, 2002
365
208
116
Which tests? At pcworld Tigerlake PL1 15W win 3 of 4 games against 4800U PL1 38W.
It's a same Lenovo Yoga Slim 7 as in Anandtech review, as seen in Anandtech it's max turbo power is 38W and sustained power level 15W.
 

IntelUser2000

Elite Member
Oct 14, 2003
7,165
1,738
136
So it could have used 38W depending on how long their gaming test lasted: https://www.notebookcheck.com/fileadmin/Notebooks/Lenovo/Yoga_Slim_7-14ARE/metrawit.png

Anandtech blindly assumes it's running at 15W which is bogus.
Yea, actually the lowest quality review is Anandtech, then Tomshardware. Based on load power consumption, the Yoga Slim 7 is at 25W and the Surface Laptop 3 is also at 25W. For the SL3 you can even see the PL1 set as 25W. They are just assuming both are at 15W.

They are noting the GPU drivers are early and they already have drivers that fixes not being able to run some games(like RDR2). That's actually not a good policy for them. They should wait. Although with Xe game updates are supposed to be much easier.

The CPU is a bit worse than expected. Overall the conclusion still can't be made. We need few retail systems for a proper comparison.
 

Hitman928

Platinum Member
Apr 15, 2012
2,914
2,453
136
So it could have used 38W depending on how long their gaming test lasted: https://www.notebookcheck.com/fileadmin/Notebooks/Lenovo/Yoga_Slim_7-14ARE/metrawit.png

Anandtech blindly assumes it's running at 15W which is bogus.
Your linked image is to laptop power measured at the wall.

Both Intel and AMD will use more than 15W for significant amounts of time, you'd have to check both to see exactly how much they were using during gaming but no one did that as far as I can see.
 
  • Like
Reactions: Tlh97

IntelUser2000

Elite Member
Oct 14, 2003
7,165
1,738
136
It's 15W tdp chip compared to 15W tdp chip,
That's a bunch of nonsense.


Load Average numbers are higher on the Yoga Slim 7 than it is on the XPS 13 7390 by 6W, and that system is using 46W PL1 on AC and 25W PL1 on DC.

15W Icelake systems like the Acer uses only 27W on load. Computerbase.de also shows testing of the Slim 7 on both 15W and 25W.
 
  • Like
Reactions: mikk

inf64

Diamond Member
Mar 11, 2011
3,060
1,770
136
Looking at AT and THG and numbers @ 15W it's very underwhelming versus Renoir. Unless people use AVX512 and only use ST workloads (where this thing can boost to 4.8Ghz), it's worse than Renoir.GPU results at 15W are equally underwhelming. If this is their reply to 2019 Zen2 design with Vega cores then they re going to be utterly destroyed by Cezanne and its successors.
 
  • Like
Reactions: Tlh97

Hitman928

Platinum Member
Apr 15, 2012
2,914
2,453
136
That's a bunch of nonsense.


Load Average numbers are higher on the Yoga Slim 7 than it is on the XPS 13 7390 by 6W, and that system is using 46W PL1 on AC and 25W PL1 on DC.

15W Icelake systems like the Acer uses only 27W on load. Computerbase.de also shows testing of the Slim 7 on both 15W and 25W.
Anandtech's power/turbo test shows it dropping to 15W long term. Perhaps Anandtech specifically set it to 15W? I guess we could ask @Ryan Smith for confirmation.

1600371484730.png
 

LightningZ71

Senior member
Mar 10, 2017
529
475
106
The results all seem to line up well with the notion that Intel allows far more power to a single core when boosting a single thread load. As we're seeing, Renoir is limited a bit more in peak power in single core scenarios. However, as a potential customer, I only care about what the final product that I can purchase and put on my table can do, not what a test platform that is not likely to be particularly balanced with its settings benchmarks like. This is a nice capabilities preview, but, it doesn't tell me a whole lot about the final product on the shelves. However, it does look like TGL is a notable improvement over ICL, and it also looks like chips down the stack will have a lot of headroom available to them. I really do want to see how the 4 core i3 performs against the ICL i7 parts.
 
  • Like
Reactions: lightmanek

IntelUser2000

Elite Member
Oct 14, 2003
7,165
1,738
136
@Hitman928 I know some of the AMD systems do weird things with power consumption.

For example:
1600371884888.png

Both Load Average and Load Maximum numbers are significantly lower than Witcher 3 Ultra numbers. It's like systems can specifically boost to 25W in gaming but stay at 15W for synthetic and CPU workloads.
 
  • Like
Reactions: mikk

IntelUser2000

Elite Member
Oct 14, 2003
7,165
1,738
136
@Hitman928 I know some of the AMD systems do weird things with power consumption.

For example:
View attachment 30039

Both Load Average and Load Maximum numbers are significantly lower than Witcher 3 Ultra numbers. It's like systems can specifically boost to 25W in gaming but stay at 15W for synthetic and CPU workloads.
What reviewers really need to do is measure system average power consumption over a long period on AC, and then under battery using a monitoring application. This will allow them to bypass all shenanigans with power measurement play and discern actual TDP more accurately. Of course it shouldn't be the sole tool. It should be used in conjunction with the HWInfo power number.

You'll see that 15W PL1 devices are at 28-34W for Load Average, and 25W PL1 devices are at low-40s. Low end dGPUs like the MX250 will add approximately 10W on top of that.
 

IntelUser2000

Elite Member
Oct 14, 2003
7,165
1,738
136
Most people posting here don’t understand that power != TDP and why a chip capped to 15W is slower than the same chip that allows momentary spikes to 30-50W.
You know the 4800U also spikes to nearly 50W right?

Power = TDP, when considering long load, which is what matters.
 

Hitman928

Platinum Member
Apr 15, 2012
2,914
2,453
136
@Hitman928 I know some of the AMD systems do weird things with power consumption.

For example:
View attachment 30039

Both Load Average and Load Maximum numbers are significantly lower than Witcher 3 Ultra numbers. It's like systems can specifically boost to 25W in gaming but stay at 15W for synthetic and CPU workloads.
Not really sure how to explain it but from that review you pointed out they show the SOC power during Witcher 3 and it is clearly not 25W. It is averaging right around 17-18 W almost the entire time which isn't uncommon for a 15W SOC to pull a couple of extra watts, especially depending on where the read out happens.

1600373968790.png

I can't explain where all the extra power is coming from unless they are reporting peak gaming power and not average, that would make sense.
 

IntelUser2000

Elite Member
Oct 14, 2003
7,165
1,738
136
@Hitman928 That's not SoC power, but GPU power.

You do have a point about it being peak power.

Nevertheless, you can see a true 15W system uses ~30W(plus or minus 10%). The Slim 7 cannot be a 15W device.

There's also no point of a battery life test on a reference system. I remember when they used to do this. The Sandy Bridge 45W system with the iGPU had great battery life.

In reality, barely anyone used the iGPU-only setup and the battery would huff and puff to get 2.5 hours, which is less than half of the 6 hours the reference system got. Actually the U systems rarely reached 6 hours.
 
Last edited:

itsmydamnation

Platinum Member
Feb 6, 2011
2,114
1,584
136
There is like a ~50pt difference between the 4800U and 4900HS is Geekbench 5 ST integer score, which means that there is no more headroom left in Renoir even if you double the power limits(in tests the U-series are limited to 30W and H/HS often reach 60W or more). Tigerlake on the other hand scores ~35% better in GB5 ST integer while pulling 50W. Granted that Intel is still not up to the mark in perf/watt, but it's ST perf is still very impressive.
sorry but your wrong,

Geechbench for ST/power is a joke it doesn't actually stress the systems at all, uses like 7 watts peak and more like 5 watt average on my 4700u in ST @ 4.2ghz.
cpu-z stress uses about 6 watts 1T
cinebench R20 uses 9 watts 1T


what does it look like if clock limits weren't imposed and it could sustain 20watt power limit for 1T, im betting its getting pretty close to 4.7ghz and cut a large amount of the deficit.

The only thing that comes close to using 15watts on a 1T is Prime small FFT which is right around 14watts @ 4.2.
Also just add i have trouble even getting geekbench 5 to run at peak turbo on my 4700u , that is how lite of a benchmark it is , the AMD boost doesn't even think its worth while........
 
  • Like
Reactions: Zepp and lightmanek

Hitman928

Platinum Member
Apr 15, 2012
2,914
2,453
136
@Hitman928 That's not SoC power, but GPU power.

You do have a point about it being peak power.

Nevertheless, you can see a true 15W system uses ~30W(plus or minus 10%). The Slim 7 cannot be a 15W device.

There's also no point of a battery life test on a reference system. I remember when they used to do this. The Sandy Bridge 45W system with the iGPU had great battery life.

In reality, barely anyone used the iGPU-only setup and the battery would huff and puff to get 2.5 hours, which is less than half of the 6 hours the reference system got. Actually the U systems rarely reached 6 hours.
Prior AMD SOCs didn't have separate sensors for GPU/CPU power, it was just SOC power. I doubt it has changed with Renoir but if you have a source for this new capability I'd be happy to be corrected.
 
  • Like
Reactions: Tlh97

Markfw

CPU Moderator, VC&G Moderator, Elite Member
Super Moderator
May 16, 2002
20,812
8,997
136
What I would like to see, is something that stresses the CPU for hours, then the firmware would have to deal with the sustained load, and the cooling solution of the hardware.. This would then show the REAL max sustained power, and how well it performed at that level.

For example, run Rosetta@home or some other DC app, even folding@home, that takes hours to complete, and then see how long it look for a specific unit, and the points awarded.

What I see here is a discussion of how long each CPU stays@ what power, and what rating it gets, but the timelines are all too limited. The longer range test would better tell the power at the wall, and how it performs. Do we really care what software says its taking for power ? Or what is it REALLY taking in worse case scenarios. And how well does it perform at that power level.
 
  • Like
Reactions: Drazick and Tlh97

IntelUser2000

Elite Member
Oct 14, 2003
7,165
1,738
136
Prior AMD SOCs didn't have separate sensors for GPU/CPU power, it was just SOC power. I doubt it has changed with Renoir but if you have a source for this new capability I'd be happy to be corrected.
You are correct. The Ryzen 3000 mobile doesn't have it. The Ryzen 4000 mobile chips do.

This is true for both HWInfo and GPU-Z, so an extra sensor was added. GPU ASIC power in HWInfo and GPU Chip Power in GPU-Z. I'm not sure exactly how that number works.

What I see here is a discussion of how long each CPU stays@ what power, and what rating it gets, but the timelines are all too limited. The longer range test would better tell the power at the wall, and how it performs. Do we really care what software says its taking for power ? Or what is it REALLY taking in worse case scenarios. And how well does it perform at that power level.
Agreed.

Software measurements are fine, as long as its reporting from the right sensors, and we understand what it means.

Wall power can be misleading, especially on battery. That's a start though, and nearly every site forgoes testing that.
 

ASK THE COMMUNITY