***Gigabyte P34W v3 "Phantasus" owner's lounge***

Does anyone know how the P34W v4 with 5700HQ compares in terms of temperature to the P34W v3? the 14nm transistors should consume less power and produce therefore less heat than the 22nm of the 47XXHQ. is that true?

 

Theoretically, yes. Practically, some real world tests are only showing a 10% improvement in temps. Can't wait for the v4 reviews to surface.

Sent from my XT1254 using Tapatalk

 

that's 8 degrees when it gets to 80°, not so bad... also a repaste with liquid ultra should get another 5 or so down. hope that makes it suitable for office use.

i'll get one the next days, probably tomorrow or Monday. i'd love to do some tests but since i have no P34W v3 here, how can i make tests/benchmarks which can be compared in an objective way to the v3?

 

Run 3dmark11 a few times and use HWmon or throttlestop to monitor max temps of GPU and CPU.

 

i got the EU model today. the one with no SSD and no windows, GA-P34Wv4-D1

i7 5700HQ
GTX 970M 3GB
1x 8GB Crucial Ballistix Sport BLS8G3N18AES4
1TB WD Blue Mobile WD10JPVX

installing win10 and the drivers from usb stick was straight forward, smart update works pretty good. latest bios is FB02.

here are some test results in a pretty warm room with 27.5° and high humidity, pretty hot these days in berlin...

idle temps:

CPU cores 49°-52°
GPU 45°

with some internet browsing:

CPU cores 63°-66°
GPU 48°

3dmark11 in gaming fan mode and balanced power setting in the gigabyte smart manager:

min: 8790
average: 8890
max: 8903

CPU cores 95°-96°
GPU 79°

3dmark11 in silent stealth fan mode and balanced power setting.

min: 5716
average: 5735
max: 5758

CPU cores 70°-72°
GPU 65°

then i noticed one pretty odd thing, the performance is worse if you set the power mode to "High Performance" in the smart manager. then the 3dmark11 dropped to:

min: 8301
average: 8440
max: 8592

but temps were a bit lower, but only 2° or so... so don't use the high performance mode in the smart manager. or is there another explanation for this behavior?

next steps are re-paste with liquid ultra and installing a Samsung 850 Evo 500GB mSATA module. a second 8GB ram is also ordered... should be a nice machine then...

only drawback is that the high res display is not yet available in the EU. but i couldn't wait since i need the machine next week for work.

 

I've just received my P34W v3 and I'm pleased with it. Coming from a Razer Blade, the biggest drawback is the trackpad (which is a big deal). Sadly, the P34W is not nearly as smooth and consistent as the Razer. Other than that, this is a great machine so far. Once you put it on the Quiet fan setting, it's not nearly as loud as some of the reviewers make it out to be.

One question: What would be the optimal SSD/HDD combo in terms of minimizing heat?

 

Interesting. Thank you for the benchmarks. Your 3dmark scores (and temps) confirm to me that the machine is still throttling. If you were to disable turbo, I think you you actually see higher performance on 3dmark as well as lower temperatures. Try it.

For comparison; at 24C ambient I get 8800 on my v3, with cpu peaking at 95. Disabling turbo, I get 8570 with CPU temps hitting 89.

My blade gets 8770 without turbo peaking at 80C. Turbo on also peaks around 94C but I get 9200+.


edit: also, can you get the "customize" fan profile to work? Mine wouldn't do anything, so I went back to smart manager v 4.

 

Comparison is up

http://www.ultrabookreview.com/8110-razer-blade-gigabyte-p34wv3/

 

that was a good hint. i chose the simple solution with 99% max CPU setting in the windows power management. this gave me 8929 in 3dmark11. but its only 20 more than the max score with turbo but tempereatures were only around 90° compared to 95° with turbo. i think i will try in the next days ThrottleStop with the settings you recommended earlier in this thread.

this is strange, the user interaction is quite unclear. i can set the fan mode to customize and then move the slider. the slider stays where it is as long as i don't click on one of the buttons 'Normal' or 'Gaming'. clicking resets the slider always to 75%.

what i tried the is clicking the 'Gaming' button and setting the slider to 100% but i had the feeling that it didn't affect the fan nor the 3dmark11 score. seems to be buggy...

 

Great review. Spot on. This will be helpful for people on the fence.

 

You have the battery capacity for the two swapped in your chart. P34W is 61.25 Wh, Razer Blade is 71 Wh.

Also, since the picture directly above the chart has the Razer Blade on the left and P34W on the right, it might be better to do the same with the chart and have the Razer Blade on the left and P34W on the right.

Edit: When talking about DPI scaling, it should be "I'm looking at you Microsoft."

 

btw, what is a good sleeve for the P34W? in europe gigabyte does not ship the notebook with a sleeve like in other countries. i am a bit annoyed... but i couldn't find the original sleeve here either. can someone have a look and tell me the part number of this one? or recommend another sleeve which fits well and has good protection? i need so move the laptop every day around in my backpack.

 

oh ****, just wanted to do a re-paste with liquid ultra and noticed that two screws of the cooling system have those evil stickers... i fear that gigabyte introduced this either for the P34W v4 model or for europe. could the v3 owners who did a re-paste confirm that these stickers are NOT on their model? damn...

 

Those are screws for the fans. You only need to take out the 6 visible ones.

@Solandri Thanks

 

phew.. thanks a lot! will do another round of tests before and after the re-paste. its much cooler today in berlin.

 

This is potentially awesome! I've not fired up my P34 for a while as I've been using the desktop a lot recently but does the customise setting work????

 

Sure doesn't lol

 

Ah ok, maybe it needs a BIOS update?!?!

 

I'm on the latest BIOS (FB04/F002) and with the latest version of Smart Manager (5.0.6) that has the customise option but can't get it to work for the life of me. Interestingly Smart Update says the latest Smart Manager version for me is 5.0.5 but when I downloaded it direct from the site for Windows 10 it gave me 5.0.6.

Either way, doesn't work properly..

 

Haha! yeah I have exactly the same. I think it's fair to say that "Smart Update" is less than useless, it's horribly slow and downloads seriously outdated drivers. I'm sure most (if not all) of us here are able to find and download the latest drivers from the respective hardware manufacturers (Intel, Nvidia etc..).

I love my P34 but I find the one thing that lets it down (throttling aside) is the quality of Gigabyte software, it feels old and clunky and quite frankly a bit s#*t.

 

At least like you said we can just install manufacturer drivers. A Sony Vaio I used to have had seriously out-dated drivers and you had to force install certain newer drivers if you weren't using their update program.

 

here are the benchmarks after the re-paste with liquid ultra, pretty impressive:

idle temps, only a bit lower:

CPU cores 48°-49°
GPU 44°

with some internet browsing:

CPU cores 55°-62°
GPU 46°

3dmark11 in gaming fan mode and balanced power setting in the gigabyte smart manager 500 points higher and 6° less heat:

min: 9474
average: 9496
max: 9530 http://www.3dmark.com/3dm11/10202769

CPU cores 88°-92°
GPU 75°

3dmark11 in silent stealth fan mode and balanced power setting, more consistent results, 180 points higher and 3° less heat:

min: 5886
average: 5900
max: 5916

CPU cores 67°-70°
GPU 62°

 

I have placed my order for the P34W v4 as well, and intend to do the repasting job with Liquid Ultra. Did you have any issues when doing the repasting? Anything I should look out for?

Also, did you see any thermal throttling or did you adjust the voltage, multiplier and disable the turbo mode to prevent it?

 

yes, i had an issue with it. at first i only applied liquid ultra to the surface of the chips (GPU, CPU) which looked pretty good. but the results were not as good as expected. i could also see that two cores had better cooling then the other two. so i've opened the machine again and applied liquid ultra to the heat sink as well (i could see some paste where the chips made contact with the sink). that was much, much better and gave the performance results above.



so my advice is:
- take some time and be relaxed, don't do it in a hurry
- clean the old paste with some paper towels and q-tips (this takes some time)
- use the alcohol cleaning sheet delivered with liquid ultra to clean all contact surfaces again and dry with another paper towel
- i've used the sponge-like thing delivered with liquid ultra to roughen the surfaces a little bit, use small circular movements and low pressure. i've used it more on the sink than on the chips.
- put a small drop of liquid ultra on the two CPU chis and a little bigger one on the GPU and spread it over the whole chip surface (also takes some time)
- then i would place the heat sink gently on the chips and remove it to see where the chips make contact
- also put a small drop on those areas and spread it to rectangular shapes roughly the size of the chips
- then slowly put the sink on the chips and try to avoid movements sideways
- fasten the 6 screws and after that put some more pressure with the fingers on the areas of the sink over the chips that you have a good initial contact
- put the cover on and do some benchmarks before you fasten all screws to see whether everything went fine

i've watched this video before:


all the best!

 

Thanks for the great tips


Sent from my iPhone using Tapatalk

 

regarding the throttling. i've not changed anything special and always used the 'Balanced' power setting in the gigabyte SmartManager. i think the CPU runs in turbo whenever it wants but does not get over 92° so it should be fine because throttling happens at 95° as far as i know. at least 96° was the max temperature before the re-paste and there was definitely throttling going on.

i've also tried the 99% max CPU load in the windows power setting to disable turbo but that decreased the 3dmark11 score by about 200-300 points. so i am hoping that the re-paste as well as the lower heat production of the 14nm 5700HQ leads to no more throttling at all...

but is there a tool which can record the CPU speed and multiplier during benchmarks? it would be interesting to see what happens and whether a specific CPU multiplier setting can gain some more performance...

 

Intel XTU is perfect for what your describing. You can undervolt as well, which will probably shave off another 3-5 degrees.

Sent from my XT1254 using Tapatalk

 

This is great news if we do not have to disable turbo mode.


Sent from my iPhone using Tapatalk

 

wow... thanks for the XTU hint, i've even turned up the multipliers to 37x which was the max XTU allowed and 2 over the default setting. now the core frequency goes up to 3,7 GHz in turbo and still no throttling. but now all 3dmark11 scores are well over 9500, the max was 9554: http://www.3dmark.com/3dm11/10209596

 

That's impressive. I think I need to enable turbo and OC my 970 to hit 9500 on my 2015 RB.

Can you tell me the temps from an extended session? (~2 hours) I try not to let GPU go above 85C and CPU over 90C peak.

 

Questions regarding temps, throttle, repaste etc...
Are there any tiny programs (windows updates, things like that), that could overheat or, on the contrary, cool down a computer? What about external devices?

I am asking because I few weeks ago I decided to do a repaste with artic silver (first time I tried it: so far I had been using a generic thermal paste). My CPU would reach 95°C and throttle quite fast one some games if I let it around 3.0Ghz, or above. (note, I always undervolt it, around -60mV)
I wish I had precisely kept track of my temps after the repaste, but I am pretty sure that I was able to play some CPU demanding games (like Grim Dawn) at around 3.0Ghz with 85°C. Quite an improvement.
Then, one day, back 93°C and above after a few minutes ingame, throttle, etc. Only difference is that on the previous day I had launched Ccleaner and removed a bunch of stuff.
Now I have to go back to my good old XTU settings and lower that CPU clock. Setting the fans to ultra high doesn't seem to change much in terms of temperature.

Those kind of inconsistencies with temperatures have been happening a lot.
Another example: The Witcher 3. That one used to always kill the CPU in terms of temperatures. In the US, I had to lower the CPU to 1.5Ghz to avoid getting near the 90°C mark.
Then I went to Vietnam for a few weeks: you an imagine how hot it was in there, specially in summer. And yet, guess what? The laptop was running cooler! I could crank up the CPU to 2.0 Ghz, even a bit more, with no problems whatsoever with the Witcher 3!
Now, I have to lower it to 1.5 Ghz again (and I'm no longer in Vietnam).

I have been wondering, could it also come from whatever devices are plugged in the computer? I use a HDMI screen, a VGA one, audio interface, mouse and keyboard. That would be worth checking (I deactivated all the screens today and only used the VGA one: didn't make any difference)...
But it still doesn't explain how the laptop seemed to run super cool for a few days after the repaste, while it still had all of these devices plugged in!
...Or maybe the thermal paste worked very well only for a couple of days?

I literally can't explain it, so all theories are welcome! ^^

 

Just bought one (coming on monday) just wondering since removing the back doesn't void the warranty and heat kills battery capacity can the battery be removed easily? I tried looking and found no disassembly videos

 

yes, i've used Heaven Benchmark 4.0 to run a longer session and the GPU went up to 81°, not 2 hours though. CPU was 92° max but usually around 89°.

i have to add, that the CPU was at 37x and GPU overclocked with +130 MHz and VRAM +650 MHz... wait what, overclocked !? 10369 http://www.3dmark.com/3dm11/10211830

 

Do you find that the VRAM increases the score much? If it doesn't increase the heat then it's fine.

 

this sound quite counter intuitive and the only guess i have might be humidity of the air. but there it gets esoteric: http://www.overclockers.com/forums/showthread.php/515952-Does-humidity-affect-air-cooling

but you could try and check the windows system timer period, maybe that has an effect, also read the linked article: http://vvvv.org/contribution/windows-system-timer-tool

and yes, backround windows updates or antivirus software can also have an effect.

it could also be dust in the hat sink, you could try to clean it and apply another bit of paste?

 

the battery is on the other side of the mainboard, should be fine in terms of heat. but it also looks like you can remove it without too much hassle:

 

Hey guys, did anyone else notice the beta bios in Smart Update (RB04T06)??

If so, anyone know what it's supposed to do?!!!

 

What elevation is your normal home at approximately? If you were at a lower altitude in Vietnam that could account for some of the difference. I've noticed some differences in my p34w v3 between my residences. One of them is around 5000 feet and one is at 0 (at the beach) my laptop allways seems to run a little cooler there I never thought about it till I saw your post.

 

I've just replaced a battery on my Gigabyte U2442N (forebear of the current model) due to significant heat degradation. It was basically useless after 2 and half years of regular use. I would think this would be the case on the current P34w model as it has similar heat production (actually slightly more comparing the two notebook check reviews), and the same internal layout.

I haven't used my laptop for a huge amount of gaming, so I would attribute most of the heat damage to it being used under moderate load (i.e. watching youtube with it on my lap, unventilated). I live in Scotland so ambeint temps are low-to-normal. I don't think my laptop has ever seen ambient temps north of 25 degrees C. Although the battery is away from the CPU and GPU, I reckon heat will still circulate everywhere within the case, especially when ventilation is hindered.

To quantify my battery being "useless", I would say it lasted about 45 mins away from the wall with just web browsing, the Windows battery icon always said I had a lot more power left, but I didn't and the laptop would suddenly cut out with no warning. So I ended up ordering a replacement battery. The replacement costs over £100 through Gigabyte - maybe there are other sources? I couldn't find any. Just be prepared to pay this every 2-3 years if you are planning on keeping this laptop for a long time - or invest in a decent cooling pad - which I might now do.

Also, yeah, it is easy to replace, at least.

 

Ooooh maybe it makes the smart-manager fan setting work? Wish I still had my P34 to test.

 

Well I went ahead and installed it last night, no problems but no extra functionality that I have found. The custom fan setting still doesn't work but it seems to have got rid of the previous settings (Stealth etc..) and now none of the options seem to do anything.

Whilst I realise the BIOS is beta, a changelog really would be helpful....

I've submitted a ticket to Gigabyte to see what they have to say for themselves!

 

For replacement batteries I found this place.... Might be a slightly better option than direct from Gigabyte (depends on shipping I suppose)

http://www.gogo-shopping.com/laptop...a013f-battery-for-g-notebook-battery-474.html

 

^Oh, yeah that would be much better value. Looking back I actually paid £144 ($227) so I would go with that store if you can! I doubt shipping will be that much.

 

The battery actually heats up a lot. If you are gaming you can feel the heat on the bottom of the case reaching the battery space. Don't forget that the mechanical harddrive also leaves a lot of heat inside the case which is located next to the battery.

 

Well I opened it up and..... got scared lol. Is the battery held in with glue? I didn't want to pry it too much so I thought I would ask here first

 

It indeed stuck with some adhesive on the back. But some where able to pry it off. I know that there are some replacement batteries on the market though.

I wouldnt risk to break the battery though. They can really erupt violently into a blazing fire. Isaw the battery health going down in 2 months though. So that was one of the reasons for me to return it to Amazon.

 

So, the guy I sold my P34Wv3 to just reported that the GPU stopped responding while playing Diablo 3, so he'll have to send it in. :/ No idea what causes these issues with the P series. I know I'm not the only one who had the GPU die on them, but 3x on the same model? What's going on?

 

Thats really unfortunate dude, there is obviously a lack of quality control. The confusing thing is that it's not consistently bad as I'm yet to have any of the many problems I have read about but GPU failures (especially 3 for the same customer!) are beyond serious.

All this makes me want to do now is benchmark the life out of mine to see if it falters..... Before my warranty runs out!

 

Hi, I have been looking to buy v3 since v4 seems to be sparsely available in Europe (I'm in the Netherlands). I can see some German models though. Where did you buy yours and what was the keyboard layout? Is it US universal?

 

i bought it directly from the german gigabyte importer simac-gmbh, i had to do some phone calls to get it. my keyboard layout is german and i have the "barebone" model without ssd and no windows and one 8GB ram module. from what the tests show i would wait to get the v4 since the 5700HQ CPU seems to produce less heat and heat is a big issue for this laptop. there are some stores which have it in stock though... you can have a look here for example: http://www.heise.de/preisvergleich/gigabyte-p34w-v4-ga-p34wv4-d1-a1294779.html

but i guess you want to have US keyboard layout...