Slashdot is powered by your submissions, so send in your scoop


Forgot your password?
Note: You can take 10% off all Slashdot Deals with coupon code "slashdot10off." ×

Comment Re:Bullshit (Score 1) 350

"Maybe someday there will be a non-Unix OS that isn't shit, but I am not holding my breath."

Personally, I'm waiting for another non-Apple UNIX that actually makes heavy interactive graphics work as smooth for the user as it is on Windows. Linux and FreeBSD both feel clunky and choppy, no matter what scheduler settings I use, or how I config X etc. IRIX was the last one I used that actually felt decent for interactive use, but they also went with the approach Windows used, and Apple later chose for OSX, with a tight graphics stack integration, from kernel up to desktop.

Comment Re:No it hasn't (Score 1) 157

Hypertransport and QPI both cause problems with communication between devices on different PCI-E switches, that's one of the things Infiniband lets you get around(and Infiniband is about so much more than just RDMA, it's also about the flexible structure you can tailor for your use, and maintain single-microsecond latencies, with all the bells and whistles. Despite the hurfblurf, RDMAoE has not made that happen).

Here's an example of the issues involved:

Comment Re: As much as possible (Score 1) 350

That's only if you think in terms of top-end movies etc. However, there's plenty of bread and butter work being done where the final rendering is done on the desktops of the artists at small studios etc, such as advertisement spots or stills, product design, architectural visualization etc etc, and that's just on the professional side. You also have to factor in the hobbyist side.

Comment Re: As much as possible (Score 4, Insightful) 350

Bullshit... Video rendering and 3D creation and rendering was desktop work already back in the 90's for hobbyists and small studios, and remains so to this day.

That's what pissed me off the most with the article, the video test was limited to encoding, not actually editing clips, working with layering, effects etc. Likewise, the Blender test is very limited in how many textures and complex multi-layer shaders are involved, it stresses geometry and rendering to a greater degree.

Not to mention that when you sit and actively work with a scene, you often have photoshop, gimp or some other program open too, as part of your workflow, creating textures, UV-maps, light maps, shadow maps, mattes&masks, height maps and normal maps etc etc...

Comment Re:No it hasn't (Score 1) 157

Hypertransport and QPI are only between CPU and either a communications hub or straight to the northbridge, and the internal shared bus(AMD and Intel have both shied away from Crossbar Switches, though HP has a custom one for their Superdome machines(which itself tries to go into Big Iron areas)). Things that would make my current Sandy Bridge Xeon or my previous Opteron system choke, such as trying to make multiple infiniband cards run at full steam simultaneously to and from a RAM disk didn't even register as a blip on system load on a mainframe when my software set was tested.

As for the offload, this isn't individual devices, but rather additions to the communications hub that give you more feature, such as hardware checksumming on the actual transfer between device and RAM, or on-board hardware crypto, essentially zero load to the CPU, and independent of any devices.

Comment Re:No it hasn't (Score 1) 157

As to the interconnect, there's also Infiniband. However, the BIG thing with mainframe I/O is that unlike the PC hardware, there's no single system bus that gets split between devices, instead you have a crap-ton of channels that can all communicate concurrently, and the available features for data integrity, encryption etc, which includes checksumming of transfers between devices or between device and RAM etc.

Comment Re:Wow! (Score 1) 99

Who says it's only about work? I do things like this for a hobby, and I'm not going to waste money by buying YET another system when it's not necessary. Besides, the 8+ core Xeons tend to sacrifice clockspeed instead, making them less useful for when you need the single-threaded performance for some tasks.

As for the link, keep in mind, that's the energy just for a single still image. Multiply for every frame in an animation as necessary, and you can see where the savings pile up.

Comment Re:And My i7-920 @ 3.8 Ghz Lives On..... (Score 1) 99

I did a crapton of 3D rendering and such back then, and AMD K6's were basically dead in the water as far as anyone doing it on a hobby level beyond mere dabbling or professionally was concerned, especially when you also had the issues with AMD's AGP support(which persisted even partially into late K7 revisions).

Comment Re:Wow! (Score 1) 99

I own Sandy Bridge systems too, and there are some reasons why I am seriously looking at upgrading. A noticeable performance increase, with quite a bit of power saving is always nice. A swedish hardware site has taken up another metric for it, by calculating the amount of joules used in total to render the scene Island in Blender at a set resolution and quality level, minus the rounded off idle energy use by the 980 Ti used in their tests. All results in joule, lower is better:

As of next Thursday, UNIX will be flushed in favor of TOPS-10. Please update your programs.