Become a fan of Slashdot on Facebook


Forgot your password?

Comment: Re:Why do these reaction wheels keep failing? (Score 1) 28

by flowerp (#48633917) Attached to: Kepler Makes First Exoplanet Discovery After Mission Reboot

Well you could secure the wheel using some rectractable bolts during launch. Only when in orbit you retract these bolts and then you turn on your electromagnets.

I found quite a lot of research papers on magnetic suspension of reaction wheels. I really wonder why this isn't mainstream technology yet.


Comment: This is not patent trolling. (Score 2) 110

by flowerp (#47833771) Attached to: NVIDIA Sues Qualcomm and Samsung Seeking To Ban Import of Samsung Phones

nVidia holds a lot of patents in the fields of graphics technology - it is a major player in this field and to date has a large market share in the desktop amd mobile GPU market. This is absolutely no patent trolling.

It's just the usual insane patent wars among major players in technology. I highly doubt this will go to court. There will just be a quiet agreement among the parties involved before this escalates too much.

Comment: I bought one of these for Litecoin mining (Score 5, Interesting) 76

by flowerp (#47772433) Attached to: Fake NVIDIA Graphics Cards Show Up In Germany

I made a test order of one of these products for evaluating whether they are any good for mining. The 4 GB video RAM on the card and the supposed graphics chip on the card would have made a very good deal.

But it became apparent immediately that this was an outdated Fermi gerneration chip, despite the card being recognized as a GTX 660 by the driver. The card ended up on my scrap heap because it was useless for my purpose (high power consumption and low performance)

At the time I assumed it was some kind of OEM product (relabeling older chips under newer product names is very common in the GPU business). But the investigation of the c't magazine seem to indicate that there is some VBIOS tampering going on and that this is not happening with nVidia's blessing at all.

I'll be following the story closely to see what the outcome of this clusterfuck will be.

Comment: tesselate polygons, then warp in vertex shader ? (Score 1) 55

by flowerp (#45213145) Attached to: Improved Image Quality For HMDs Like Oculus Rift


Can't one just subdivide (tesselate) polygons that appear relatively large in screen space so that they consist of many small polygons, with a few pixels each? This would allow for doing the barrel distortion entirely in the vertex shader with no ray tracing being required. The challenge is to perform the dynamic tesselation without requiring a constant updating of the geometry (vertex buffers) on the GPU.
Maybe newer APIs like DirectX10 or 11 would support this dynamic tesselation approach in hardware.

Comment: 1000.2 TFLOPS reached! (Score 1) 96

by flowerp (#42459703) Attached to: Einstein@Home Set To Break Petaflops Barrier

I added two nVidia GTX 260 and one nVidia GT 240 card to Einstein @ Home , and voila this morning's stats show:

Page last updated 3 Jan 2013 8:50:02 UTC
Floating point speed (from recent average credit of all users) 1000.2 TFLOPS

For a BOINC novice it can be quite daunting to figure out how to make it use all GPUs and not accept any CPU-only work units. Editing some XML files in some data directory isn't exactly user friendly.

Comment: Re:too specialized on a single protocol? (Score 0) 357

I did not miss anything here.

If I send a video stream as a sequence of UDP or RTP packets, clumping together to perform some kind of forward error correction is perfectly possible and reasonable.

When you invent some kind of solution to prevent packet loss on wireless links, it should apply to all kinds of IP traffic and not single one protocol.

"If anything can go wrong, it will." -- Edsel Murphy