5 ms·

Yeah well I doubt they keep that stance for very long. Nvidia is investing like crazy in keeping the deep learning community happy about their products and they

by Iv 9y ago

Yeah well I doubt they keep that stance for very long. Nvidia is investing like crazy in keeping the deep learning community happy about their products and they are disproportionately on Linux. If/when Wayland becomes the default for Ubuntu, Nvidia will follow. Can't let them ponder the amount of time it would take to code the missing layers for AMD hardware

mkj 9y ago
A bit off topic, but do the people using deep learning GPUs use the same device for desktop graphics? There aren't problems with display lockups etc?
- sdenton4 9y ago
  My home deep learning machine only had a monitor long enough to install Ubuntu. After that it's been accessed entirely through ssh on my home lan... I sometimes fantasize about hooking up twelve monitors for a wicked game of descent, though.
- n17r4m 9y ago
  I can only speak from my own experience (2-3 GPU rigs running tensorflow) but I have never had any trouble. Full Ubuntu desktop. Training models on the GPU does not cause any problems with display lockups etc.
- freeone3000 9y ago
  We don't use the display at all - there's not a lot of space in the server room for monitors.
- danieldk 9y ago
  We can't even if we wanted too ;) (most Teslas don't have video output).
- rprospero 9y ago
  Not exactly on the deep learning side, but I am a scientist who does some computatinoal work on the GPU. You're right that we don't use our computer card for our graphics and, as many people pointed out, the computation will wind up running on a headless server. On the other hand, during development, our code will be running on our desktops and will be using the GPUs on the graphics card. I've witnessed a developer's frustration with X11 driver issues cause a project to transition from CUDA to OpenCL. Granted, he didn't just throw a hissy fit an port the code. Rather, it inspired him to write up a tradeoff analysis of OpenCL versus CUDA that showed that OpenCL offered this specific project significant benefits and that the main selling points of CUDA weren't being used in this instance. He also wound up rage coding a prototype port to OpenCL to go along with the proposal. The project would have moved eventually, since it was the better engineering decision in this case, but it was about a half decade away on the timeline until the lead dev's compositor stopped work.
saghm 9y ago
Wayland is the default for Gnome now, and Ubuntu is switching to Gnome for its desktop next year. Have they said anything about whether they plan to change the default from Wayland to X for their installations?
padraic7a 9y ago
At one stage Ubuntu seemed to be definitely planning a move to Wayland. Because of it's performance however they may be pulling back from that: http://www.omgubuntu.co.uk/2017/07/ubuntu-uncertain-using-wayland-default http://www.omgubuntu.co.uk/2017/07/ubuntu-uncertain-using-wa... As a Fedora user I can understand the reluctance. The recent upgrade to Fedora 26 has given me a ton of gnome crashes which I think are Wayland related. Commentary on Fedora channels tends to be along the lines of "Wayland is great and works for everyone (OK, maybe not those with NVidia cards)". If Ubuntu has millions of users it may have millions of NVidia users. Putting them onto unstable technology to pressure Nvidia into supporting Wayland wouldn't be a great move.
- Iv 9y ago
  I say if/when because I have read about Ubuntu plans to switch for at least two years. Therefore I can understand Nvidia's lack of haste. They will support the version/fork that Ubuntu chooses. No point in wasting resources until then. Of course that would not happen if they opensourced their drivers but I kinda lost hope about that...
ekianjo 9y ago
> If/when Wayland becomes the default for Ubuntu Ubuntu is the default distro for gaming, too, and I doubt they will jump until it's well ready, and certainly NOT before nVIdia provides support. It would make no sense.
- majewsky 9y ago
  Then again, Ubuntu has a long tradition of not making sense (e.g. how they pushed PulseAudio onto users before it was ready for primetime).

by Iv 9y ago

mkj 9y ago

A bit off topic, but do the people using deep learning GPUs use the same device for desktop graphics? There aren't problems with display lockups etc?

sdenton4 9y ago
My home deep learning machine only had a monitor long enough to install Ubuntu. After that it's been accessed entirely through ssh on my home lan... I sometimes fantasize about hooking up twelve monitors for a wicked game of descent, though.
n17r4m 9y ago
I can only speak from my own experience (2-3 GPU rigs running tensorflow) but I have never had any trouble. Full Ubuntu desktop. Training models on the GPU does not cause any problems with display lockups etc.
freeone3000 9y ago
We don't use the display at all - there's not a lot of space in the server room for monitors.
danieldk 9y ago
We can't even if we wanted too ;) (most Teslas don't have video output).
rprospero 9y ago
Not exactly on the deep learning side, but I am a scientist who does some computatinoal work on the GPU. You're right that we don't use our computer card for our graphics and, as many people pointed out, the computation will wind up running on a headless server. On the other hand, during development, our code will be running on our desktops and will be using the GPUs on the graphics card. I've witnessed a developer's frustration with X11 driver issues cause a project to transition from CUDA to OpenCL. Granted, he didn't just throw a hissy fit an port the code. Rather, it inspired him to write up a tradeoff analysis of OpenCL versus CUDA that showed that OpenCL offered this specific project significant benefits and that the main selling points of CUDA weren't being used in this instance. He also wound up rage coding a prototype port to OpenCL to go along with the proposal. The project would have moved eventually, since it was the better engineering decision in this case, but it was about a half decade away on the timeline until the lead dev's compositor stopped work.

saghm 9y ago

Wayland is the default for Gnome now, and Ubuntu is switching to Gnome for its desktop next year. Have they said anything about whether they plan to change the default from Wayland to X for their installations?

padraic7a 9y ago

At one stage Ubuntu seemed to be definitely planning a move to Wayland. Because of it's performance however they may be pulling back from that: http://www.omgubuntu.co.uk/2017/07/ubuntu-uncertain-using-wayland-default http://www.omgubuntu.co.uk/2017/07/ubuntu-uncertain-using-wa... As a Fedora user I can understand the reluctance. The recent upgrade to Fedora 26 has given me a ton of gnome crashes which I think are Wayland related. Commentary on Fedora channels tends to be along the lines of "Wayland is great and works for everyone (OK, maybe not those with NVidia cards)". If Ubuntu has millions of users it may have millions of NVidia users. Putting them onto unstable technology to pressure Nvidia into supporting Wayland wouldn't be a great move.

Iv 9y ago
I say if/when because I have read about Ubuntu plans to switch for at least two years. Therefore I can understand Nvidia's lack of haste. They will support the version/fork that Ubuntu chooses. No point in wasting resources until then. Of course that would not happen if they opensourced their drivers but I kinda lost hope about that...

ekianjo 9y ago

> If/when Wayland becomes the default for Ubuntu Ubuntu is the default distro for gaming, too, and I doubt they will jump until it's well ready, and certainly NOT before nVIdia provides support. It would make no sense.

majewsky 9y ago
Then again, Ubuntu has a long tradition of not making sense (e.g. how they pushed PulseAudio onto users before it was ready for primetime).