NVIDIA Releases $15,000 USD Linux-Powered Developer System

[deppman]

how to solve the display problem

Potential buyers will love compiz crashing, and this https://devtalk.nvidia.com/default/t...-12-gtx-580/1/

This solves the problem:

$ sudo install kubuntu-desktop

And then use KDE, which displays perfectly. When I buy a System 76 system with stock Ubuntu, its the first thing I do.

OTOH, it appears that NV may have worked with the Compiz maintainers to remove at least some of the bugs from Compiz. See http://www.fudzilla.com/news/graphic...-old-linux-bug

[droidhacker]

This is *NOT* a "dev" box. This is .... mayyyyyybe a research box. At best.

One of the characters for a DEV box, is that it should *NOT* be inconsistent with your target (consumer, business, other) box, otherwise how software will work on the dev box will not be an accurate reflection of how end users will have it work.

In other words, dev box should be under $1000 in most cases.

[marciocr]

Potential buyers will love compiz crashing, and this https://devtalk.nvidia.com/default/t...-12-gtx-580/1/

Compiz? When the GNOME started using Compiz? The composer was not the Mutter?

[tuke81]

Compiz? When the GNOME started using Compiz? The composer was not the Mutter?

This does not use gnome, it uses unity and thus compiz. Not sure about is that bug in compiz or is it special to mutter, I haven't seen it with kwin.

[vadix]

Ubuntu supports nVidia's proprietary drivers with packages in their main repos. Aside from that, Linux is Linux: there's no compelling reason not to use a distro with the largest desktop footprint (and therefore largest community support) that also offers paid support.

Larger footprint doesnt preclude better support. Also, it depends on the sort of support you need. For me the Arch wiki provides far more support than ubuntu forums and wiki combined.

[dragorth]

This is *NOT* a "dev" box. This is .... mayyyyyybe a research box. At best.

One of the characters for a DEV box, is that it should *NOT* be inconsistent with your target (consumer, business, other) box, otherwise how software will work on the dev box will not be an accurate reflection of how end users will have it work.

In other words, dev box should be under $1000 in most cases.

You're missing the point of this box. First, Nvidia is trying to sell more GPUs. So, they make the Dev box with 4 of their GPUs. Because the biggest issue they have is how hard it is to take advantage of multiple GPUs in a well scaling manner. So people don't bother, they only install 1 GPU, and work from that. Unless you are buying a supercomputer, in which case you are hiring specialists to handle that complexity for you.

This sets a baseline for those that would benefit from such a system to get good at using their GPUs. This is important for them to be able to move into the supercomputer arena, as well as preparing the way for Nvidia's next GPU rollout, that can have 8 GPUs in 1 system. That is their end-game for these types of cards. They have been trying to convince us of the need of multiple headless GPUs for the last two generations, and this is a way to continue.

So consider this as Nvidia's aspirational Dev machine, because they want the actual machine to be much higher spec.

[DeeK]

Finally ...

Excellent! I've finally found the perfect system to play nethack.

[curaga]

Excellent! I've finally found the perfect system to play nethack.

By that you mean the system plays Nethack

[DDF420]

This is *NOT* a "dev" box. This is .... mayyyyyybe a research box. At best.

One of the characters for a DEV box, is that it should *NOT* be inconsistent with your target (consumer, business, other) box, otherwise how software will work on the dev box will not be an accurate reflection of how end users will have it work.

In other words, dev box should be under $1000 in most cases.

Deep learning neural networks learn many levels of abstraction. They range from simple concepts to complex ones. This is what puts the ?deep? in deep learning. Each layer categorizes some kind of information, refines it and passes it along to the next.

Deep learning lets a machine use this process to build a hierarchical representation. So, the first layer might look for simple edges. The next might look for collections of edges that form simple shapes like rectangles, or circles. The third might identify features like eyes and noses. After five or six layers, the neural network can put these features together. The result: a machine that can recognize faces.

GPUs are ideal for this, speeding a process that could otherwise take a year or more to just weeks or days. That?s because GPUs perform many calculations at once?or in parallel. And once a system is ?trained,? with GPUs, scientists and researchers can put that learning to work.

That work involves tasks once thought impossible. Speech recognition is one application. So is real-time voice translation from one language to another. Other researchers are building systems that analyze the sentiment in social media conversations.

We?re just scratching the surface. That?s why researchers at top universities worldwide and a host of startups are rushing to put deep learning to work.

Deep learning is a subset and more focused area of Machine Learning

[vadix]

By that you mean the system plays Nethack

This.