***The Official MSI GT80 Titan Owner's Lounge***

I agree with both of you. As Kevin said, nvidia will most likely address this and relatively soon because there are news about it all over the place.

But for now, I am happy enjoying amazing performance!

 

I'm always that guy who gets the under spec's cpu because I know the performance gains are near identical gaming wise but there's always that bug in my behind that haunts me wishing I went with the other option!! Sadly 009 is not instock and back ordered from my vendor

 

Seriously? I guess I wasn't aware that you were comparing the GPUs of the CPU? Why would anyone game on the Intel GPU when they have 980m cards in SLI?

I was talking about the CPUs in a game like BF4. It will be the difference of a few frames at best.

 

I know I never use the iGPU

 

Yep there wont be TOO much of a difference between the two for now.

 

That's pretty humorous. ^^

 

Thank you!
It seems there's a bug because only 61 segments of 128MB are been tested. That's 7.8GB. So, there's 1 missing segment.
How much VRAM is reported by GPU-Z?
Nonetheless, the bandwidth of the last segments (56-60) are all slower than the regular ones.

 

There's no driver solution for that problem.
We all got scammed by a multi-billion-dollar-global-corporation that's still propelling false advertisements about their junk.
<PLACE-HOLDER> NVIDIA!!!
I'll be joining the class actions lawsuits ASAP.

 

The main reasons why anyone would play on an IGP instead of a two-way 980M SLI is because:

1. It saves energy. Assuming a single 980M have a TDP of 100W and a two-way SLI usually spend 60% more[1]. The total power needed to run it full speed is 160W. An IGP shares the same TDP of the CPU and for Mobile Core i7 processors it's no more than 47W (57W for the MX version). So, the SLI will spend around 340% more power than the IGP at full speed.
   
   
2. It's good enough. The observed FPS in games are much higher with the SLI, but for certain games (low to medium demanding ones, like Dota 2) there's no observable benefit (for normal eyes like mine) using the SLI when compared to IGP using the embedded display in native resolution (1920x1080).

For those who really move with the laptop, playing the aforementioned games using the IGP will certainly give you more play time.

[1] http://www.guru3d.com/articles-pages/geforce-gtx-970-sli-review,4.html

 

That's about the bumpgate (2008). Let's see what will be their behavior this time.

 

Would anyone know if the bios on the GT72 is unlocked or allow the "Turbo Boost Power Time Window" to be changed?

 

You will need unlocked BIOS to do that.

 

In SLI the power consumption will be closer to 180-190W even if the performance is not there due to overheads.

I agree for 2d and older titles. For slightly more modern ones then using a single card in non SLI and maybe limiting the clocks a bit may work too.

 

Indeed! I'm curious about that option. Have you tried it? How easy it is to disable one of the GPUs on the GT80? Can you choose wich one will be active?

 

I've just received my replacement 001 with copper upgrade, doing some testing on temps with stock firmware and updated firmware and stock paste vs liquid ultra will post results in abit

 

Testing is almost done!

 

Have you ever tweaked anymore i want to run afew benchmarks with your tweaked settings for temp results aswell

 

So i'm not sure how well of a job liquid ultra did but i'll let you guys decide, something doesn't seem right with the firestrike temps.

in between each run was given 2 minutes of Cooler Boost Button and 2 minute idle without Cooler Boost enabled to normalize temps.

Valley Directx 11, Ultra, x8 AAx 1080p
Ambient 22c
Stock paste and stock Firmware GPU 1 (67c) GPU2 (83) CPU (89,85,87,84)
Stock paste and 103 Firmware GPU 1 (66) GPU 2 (84) CPU (89,87,85,87)
Liquid Ultra and 103 Firmware GPU 1 (71) GPU 2 (80) CPU (84,77,78,75)

3DMark Firestrike
Ambient 22c
Stock paste and stock Firmware GPU 1 (68) GPU 2 (80)
CPU (90,88,89,86)
Stock Paste and 103 Firmware GPU 1 (68) GPU 2 (81) CPU (91,89,90,86)
Liquid Ultra and 103 Firmware GPU 1 (74) GPU 2 (79) CPU (91,87,90,84)

can anyone make sense of this? as far as i know the past job was pretty spot on, i can try redoing the paste, i just dont know how easy it is to remove liquid ultra and start again, or perhaps i could add more liquid ultra, although i did make a thin layer on both the Die and the blocks
Gonna do afew more tests with -60mv undervolt and such.

 

Ya that's puzzling for sure so I would go back and inspect things again. One thing that looks evident is the cpu temps changed or moved around between cores if they were all typed in the same order consistently. This makes me feel something is not making the same contact as it did before.

As you have probably seen the cpu pipes are tucked in pretty good. There are also little wire rest mounts for the pipes so pay special attention to them if you didn't already just to ensure they aren't causing any unwanted force or resistance and things are in place and resting as they should be.

I would also apply a bit more CLU on the cpu die, I had a similar experience with my AW18 being afraid I would use too much and I just didn't use enough. After going back I was able to get the results I was looking for.

If you clean CLU it's very easy to remove with Alcohol just pay attention to ensure nothing rolls into places it shouldn't be.

Also note I only used CLU on the CPU and not the GPU so I could preform other tests incase you are referring to my testing.

 

Just curious, how do you mod your copper? By set up extra pipes or other way?

 

@GTVEVO:

Just for clarification, when you posted HERE and HERE, were you using the stock fan profile, or did you use Cooler Booster 3 during the benchmarks?

 

Regular profile in all my tests, no cooler boost engaged.

 

With all this talk of thermal compounds over another, I think Tom’s Hardware conclusion of thermal paste is most accurate…. That it is the APPLICATION of the product, that is most important:

Thermal Paste Comparison, Part Two: 39 Products Get Tested

I am about to Fire Strike test my notebook (highest CPU temps reported yet) in 'Sport Mode' with Cooler Booster engaged, and see where this places my CPU temp. After that, I'll probably re-do the TIM application.

 

Ok, the following are my Fire Strike results in Sport Mode, with Cooler Booster 3 fully engage. In my previous Fire Strike benchmark, without Cooler Booster, the CPU temp peaked to 95C..... so Cooler Booster is reducing the CPU peak temp by 12C. Also, to see the impact SLI is having on the CPU temps, I ran the same benchmark with the same parameters, using only a single GPU.

@Cormogram: Yes, you can disable SLI in the NVCLP, but GPU's (or CPU) can only be selectable for PhysX purposes. When disabling SLI, three of the start-up appz must be closed..... SCM, SB Cinema, and the Gaming Center.

Enjoy!




And below are the CPU temps, with SLI disabled, in Sport Mode, with Cooler Booster engaged.

 

Hi guys, I followed Cormogram here from the P35X thread, so sorry if I'm butting in here. We also have the GTX980M 8GB but don't have the same slow last segment phenomenon and I'm wondering why Not like it matters but it's weird.

Link to my results http://m.imgur.com/9CxPNNr

 

Thank you! That's so cool! I mean low temperature by that.
When you disable SLI both GPUs still operational?
If only one is active (draining power) which one is selected by NVCLP? Is there a main/master one? Are both of them working in x16?

 

No slow last segment because there are no partially disabled ROPs in the mobile series. The sole card to have issues is the desktop 970 which is a relief.

 

Except that's exactly what Ryzeki's test showed and what owners of the 980m 4gb version are reporting.

 

That's why I'm asking, because Ryzeki seems to have a GTX980M 8GB? I don't know why mine would be fine and not his.

 

It was already reported that the test is only accurate with optimus machines and as meaker said, the mobile gpus are unaffected. Only the desktop gpu gtx970 is affected. The software is useless in determining the memory bug.

 

Right, thank you for the clarification. I didn't think the 980M was affected but people keep insisting that it is. I'll be able to point them in the right direction now.

 

You are welcome! The issue actually exploded with conspiracy and all haha but reallythe good news is that all current mobile gpus are unaffected.

 

Basically when the GPU is dedicated only windows reserves some of the ram so the program can't actually acess that bit, windows will give it up if a program is coded properly.

 

Can I choose which Cherry switch or am I limited to red?

 

Brown (assuming you mean click level and not backlight color)... but you might be able to swap them out. Google/YouTube or maybe search this thread for a link to a video of someone swapping the guts of each key out separately.

Homer

 

So cherry brown switch is the only switch that comes with the laptop?

 

So far that is correct.

 

After lurking here since the Titan's release (I prefer to think of it as watchful waiting) I just placed my order for a model 001 from ExcaliberPC via Amazon. The most exciting thing to me is the keyboard; no one seems to have had trouble adjusting to its location, which was my main reservation. So by the middle of next week I should be "in the club" (which has a very different meaning in Britain I understand).

 

You wont be disappointed at all, tallan. Adjusting for the keyboard placement took maybe a day or two for me. After fixing the little issue I had I'm very happy with this laptop. Currently just playing Dying Light with no issues.

 

I felt the same way and doing the exact same

 

Anyone figured out about mapping macros to the keyboard, my understanding is with the steelseries this is possible with any key just about?

 

Install the latest Steelseries Engine 3! Recognizes and works great with our keyboard.

User Guide

 

Has anyone tried installing or modifying the leaked Asus G751 drivers on our gt80? It is also Connected thru eDP

 

Any idea of the specs of our sound systems as far as the build in headphone amp?

 

Has this thread been abandoned!! Lol

 

It's the weekend!

 

As striker said, its the weekend haha, had other matters to attend, though I am still testing more with my machine

And! There aren't that many owners compared to other popular msi machines yet

I am working in my review as well

 

Any answers for my previous questions?

 

Any of you guys tried Photoshop, video editing or motion graphics applications on this Jesus DTR yet? In case you don't have Adobe CC, there is a free trial and if you want to try some test benchmarks I've posted a few here along with a shared google doc where you can add your own results: http://forum.notebookreview.com/thr...iting-motion-graphics-3d-modeling-cad.769866/

Very interested to know whether the second card helps with some of the applications that can make use of it.

 

I don't think adobe's apps utilize second gpu's just yet. But I would love to see the performance of this incredible machine in Premiere Pro CC and After Effects CC as they are capable of stressing components for prolonged durations during renders.