Real Users' Benchmarks - Rank of PCS Laptops

Macco26

Macco26

Expert
1612954715757.png

Those are the 2 Elimina compared.

@Style evo I can see your CPU pushing a lot more frequency (4789 Mhz) giving so less headroom to the GPU to clock high (average 1623 vs 1669). Probably high temperature is due to the heated CPU sharing the same heat pipes of the GPU.
Also, these new RTX 30 and their Dynamic Boost 2.0 definitely like to have GPU at full potential, and CPU at no more than 35W...

I'd consider to try another performance level in the Control Center maybe. If automatically this does not change, consider trying to set a given PL1 to the CPU (like 35W or 45W max) with Throttlestop and repeat the test. Curious how it turns out.
By the picture here, it seems the less power the CPU gets, the better the Firestrike results is..

Why it differs from the other Elimina, I actually don't know... That's why you should check which Control Center setting you both used.
 
S

Style evo

Active member
I used the performance mode and a custom fan profile (such that the fans should have been at 100% at those temps). Not yet using a cooling pad but given this machine runs so hot I will probably get one to help longevity! I will try it later on entertainment mode.
 
Macco26

Macco26

Expert
I suspect (and you can get a precise idea by watching HWInfo64) that Perfomance mode has too much high PL1 limit for CPU, a thing simply not needed for Firestrike. It steal juicy watts from the starving GPU and Dynamic Boost can't happen. Keep us updated.
 
FerrariVie

FerrariVie

Super Star
Macco26 said:
I suspect (and you can get a precise idea by watching HWInfo64) that Perfomance mode has too much high PL1 limit for CPU, a thing simply not needed for Firestrike. It steal juicy watts from the starving GPU and Dynamic Boost can't happen. Keep us updated.
Click to expand...
I agree with your observations, @Macco26 . Before Dynamic Boost 2, the higher the CPU performance, the better your physics results and final score overall. However, now with the 3000 series if you have a slower CPU (less power hungry), it will give more power to the GPU and provide you better results not just in firestrike, but gaming in general. That's why I assumed that DTR's will lose place in the gaming environment, since they won't be able to use dynamic boost at all. However, you could probably allow 115W all the time on a DTR 3060, so it would be better than 90W+15W anyway :D

@Style evo , I've updated the ranking with your results, but I can replace it whenever you get better results. Do try using entertainment to check the difference and give it a try with a cooling pad later (is it cheating? :D just joking).

I would also try leaving the default fan curve in another run, just to check if it makes any difference. I don't think so, since it's running quite hot and fans will very likely be running quite fast regardless.

Edit: I've looked at the detailed results and indeed your score is below average for your specs and you were below YogiC75 in all areas (CPU, GPU and temps). Did you have any other software open at the time? Something that could be consuming resources? Even Hwinfo or sensor overlays (like Rivatuner) can lower the scores a bit. Also leave Control Center only in the tray as well, not opened on the foreground.
msedge_ovL6ceBA00.png
 
Last edited:
S

Style evo

Active member
I had no other software open other AV, google backup and sync (Google Drive) and stuff like that minimised in the tray.

I tried on entertainment mode on both automatic fan profile and my custom profile. Both came out with lower scores. I have linked my second run with my custom fan profile that definitely keeps the system cooler than automatic.

www.3dmark.com

I scored 16 726 in Fire Strike

Intel Core i7-10870H Processor, NVIDIA GeForce RTX 3060 (notebook) x 1, 32768 MB, 64-bit Windows 10}
www.3dmark.com www.3dmark.com

I think there is definitely some throttling going on - particularly noticeable in the CPU test with a sudden drop in frame rate 3/4 through as the CPU gets so hot:

1612966398856.png
 
Scott

Scott

Behold The Ford Mondeo
Moderator
Style evo said:
Shouldn’t be the paste, I went for the premium option.
Click to expand...
Depends how well it was applied unfortunately. It can be a bit hit or miss (mostly miss in my experience personally). I tend to recommend just sticking with the regular paste as it's pre-applied. The last information I had stated that employees were recommended to use the whole tube on installs to ensure a proper coverage on laptops, that's just the guidance they are following.
 
Macco26

Macco26

Expert
Scott said:
Depends how well it was applied unfortunately. It can be a bit hit or miss (mostly miss in my experience personally). I tend to recommend just sticking with the regular paste as it's pre-applied. The last information I had stated that employees were recommended to use the whole tube on installs to ensure a proper coverage on laptops, that's just the guidance they are following.
Click to expand...
I've opted for MX-4 myself for my upcoming laptop. I know this means a PCS employee may screw things up additionally, but I did this because it seems MX-4 can last a lot more (up to 8 years) and I don't want to dismount the entire heat pipe system every year if I can avoid doing so for a mere 14 eur... Hope in the good hands of PCS guy [emoji16]
 
Scott

Scott

Behold The Ford Mondeo
Moderator
Macco26 said:
I've opted for MX-4 myself for my upcoming laptop. I know this means a PCS employee may screw things up additionally, but I did this because it seems MX-4 can last a lot more (up to 8 years) and I don't want to dismount the entire heat pipe system every year if I can avoid doing so for a mere 14 eur... Hope in the good hands of PCS guy [emoji16]
Click to expand...

There's not a snowballs chance in hell that the paste will last for 8 yrs in a laptop. I think mine was around 6 months before I started getting rough thermals. I then opted for liquid metal which lasted around a year. I'm now back to MX-4 due to it being cheaper but it's starting to run quite toasty again (think I changed it around 8 months ago).

That's with a delid as well.
 
Macco26

Macco26

Expert
Eh I know inside the laptop it won't last anywhere near like their warranty Desktop as it might be set for Desktop scenarios. The 'up to' they put is for a reason :cry:
Still I do hope the MX-4 might last just a bit longer than the stock paste. Only in my dreams? We'll see. And I hope PCS should be pretty capable at applying MX-4 anyway, as they produce like hundreds of Desktop PC a day. Hope the case they apply it badly is a bad lemon happening once in a rare occasion...
 
Scott

Scott

Behold The Ford Mondeo
Moderator
Macco26 said:
Eh I know inside the laptop it won't last anywhere near like their warranty Desktop as it might be set for Desktop scenarios. The 'up to' they put is for a reason :cry:
Still I do hope the MX-4 might last just a bit longer than the stock paste. Only in my dreams? We'll see. And I hope PCS should be pretty capable at applying MX-4 anyway, as they produce like hundreds of Desktop PC a day. Hope the case they apply it badly is a bad lemon happening once in a rare occasion...
Click to expand...
They're 3 for 3 with my laptop. Like I said, when I queried I was told they are instructed to use the whole tube. Hopefully that has been addressed though :)
 
B

Besniq82

Member
Thanks I was really surprised, that the laptop finally performed the way suppose to. Just for information. Laptop was on performance mode, ambient temp was may be around 22 degrees, laptop was just positioned on a table without any extra cooling.Im using Thermal Grizzly Kryonaut. I put fresh one few weeks ago, because I purchased the laptop with i5 10400k by the way I was getting higher temps with it with the stock thermal paste from PC specialist, but with grizzly kryonaut temps are better on the i9 10850k .A also when I swapped to i9 cpu I had to reinstall the control center. The old was seems to bug with the new i9, in any mode from the control centre was sitting at 4.2 ghz. By the way I haven't overclock with Intel XTU, just the cpu adjusts its clocks depending on which mode I switch in the control center
 
M

maxff9

Bronze Level Poster
maxff9 said:
I repeated the test. Entertainment power mode. All stock.
February, outside temperature -1°C (laptop stayed in my car all day long).
I don't why the results are less good than last time ?

Chassis and SpecFirestrike
3DMark		GraphicsPhysicsAvg. GPU
Temps		URL
* Nova 15" / R5 3600 / RTX 207017530195891845767°Link
Click to expand...

Entertainment mode with changing the freq(volt) curve of the GPU.
Chassis and SpecFirestrike
3DMark		GraphicsPhysicsAvg. GPU
Temps		URL
* Nova 15" / R5 3600 / RTX 207017901201241865966°Link
* Nova 15" / R5 3600 / RTX 2070 @
1620 at 844 mV		17092189091864863°CLink
 
Last edited:
L

leebee2110

Bronze Level Poster
Another Nova to the mix...

www.3dmark.com

I scored 18 775 in Fire Strike

AMD Ryzen 7 3700X, NVIDIA GeForce RTX 2070 (Notebook refresh) x 1, 32768 MB, 64-bit Windows 10}
www.3dmark.com www.3dmark.com

Chassis and SpecFirestrike
3DMark		GraphicsPhysicsAvg. GPU
Temps		URL
* Nova 15" / R7 3700x / RTX 207018775203332346464°Link

@FerrariVie noted my graphics score trailing behind yours. I held slightly higher average clock speed but higher temp. and lower average memory clock speed (assuming temp. related.) Maybe that's the downside of the R7 :(
 
FerrariVie

FerrariVie

Super Star
leebee2110 said:
Another Nova to the mix...

www.3dmark.com

I scored 18 775 in Fire Strike

AMD Ryzen 7 3700X, NVIDIA GeForce RTX 2070 (Notebook refresh) x 1, 32768 MB, 64-bit Windows 10}
www.3dmark.com www.3dmark.com

Chassis and SpecFirestrike
3DMark		GraphicsPhysicsAvg. GPU
Temps		URL
* Nova 15" / R7 3700x / RTX 207018775203332346464°Link

@FerrariVie noted my graphics score trailing behind yours. I held slightly higher average clock speed but higher temp. and lower average memory clock speed (assuming temp. related.) Maybe that's the downside of the R7 :(
Click to expand...
Thank you! You actually got the best score overall for a Nova laptop until now, but I am indeed on the top of the graphics score for the 2070. It is definitely temperature related, as the R9 sample is worse than you and the R5 a bit better.

I'll update the main rank soon.
 
You must log in or register to reply here.
Top