Search
Search titles only
By:
Search titles only
By:
Home
Forums
New posts
Search forums
What's new
New posts
Latest activity
Members
Current visitors
Κανονισμός Λειτουργίας
Σωματείο AVClub
Log in
Register
Search
Search titles only
By:
Search titles only
By:
New posts
Search forums
Menu
Install the app
Install
Reply to thread
Home
Forums
Δικάναλος Ήχος
Do It Yourself
Διάφορες Kατασκευές
FIR @ GPU
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
Message
<blockquote data-quote="tmjuju" data-source="post: 1417713" data-attributes="member: 2204"><p>FIR @ GPU </p><p></p><p>Ένα από τα πρώτα Papers που βρίσκω που βασίζονται σε μοντέρνες υλοποιήσεις GPU. Εννοείται ότι η δουλειά του Koon ( <a href="http://www.avclub.gr/forum/showthread.php?t=29204&page=2&highlight=koon" target="_blank">http://www.avclub.gr/forum/showthread.php?t=29204&page=2&highlight=koon</a> ) στο diy δεν έχει θέση σε ένα ακαδημαϊκό paper <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite7" alt=":p" title="Stick out tongue :p" loading="lazy" data-shortname=":p" /> , αλλά είμαι σίγουρος ότι σταδιακά θα δούμε και άλλα papers. </p><p></p><p></p><p>HIGH-PERFORMANCE REAL-TIME FIR-FILTERING USING</p><p>FAST CONVOLUTION ON GRAPHICS HARDWARE</p><p></p><p>Institute of Technical Acoustics,</p><p>RWTH Aachen University</p><p>Aachen, Germany</p><p><a href="mailto:fwe@akustik.rwth-aachen.de">fwe@akustik.rwth-aachen.de</a></p><p>Institute of Technical Acoustics,</p><p>RWTH Aachen University</p><p>Aachen, Germany</p><p></p><p></p><p><a href="http://dafx10.iem.at/proceedings/papers/WefersBerg_DAFx10_P76.pdf" target="_blank">http://dafx10.iem.at/proceedings/papers/WefersBerg_DAFx10_P76.pdf</a></p><p></p><p>ABSTRACT</p><p>In this paper we examine how graphic hardware can be used</p><p>for real-time FIR filtering. We implement uniformly-partitioned</p><p>fast convolution in the frequency-domain and evaluate its performance</p><p>on a NVIDIA GTX 285 graphics card. Motivated by audio</p><p>rendering for virtual reality, our focus lies on large-scale realtime</p><p>filtering with a multitude of channels, long impulse responses</p><p>and low latencies. Graphics hardware has already been used for</p><p>audio signal processing — including FIR and IIR filtering with</p><p>respect to offline and real-time processing. However, the combination</p><p>of GPU computing and real-time conditions leads to a</p><p>number of challenges that have not been reviewed in detail. The</p><p>new contribution of this paper is an implementation and detailled</p><p>analysis of a frequency-domain fast convolution method on GPUs.</p><p>We discuss specific problems that emerge under real-time conditions.</p><p>Our method allows to achieve an outstanding real-time filtering</p><p>performance. In this work, we do not only regard a timeinvariant</p><p>filtering, but also time-varying filtering, where filters are</p><p>exchanged during runtime. Furthermore, we examine the opportunities</p><p>of distributed computation—using CPU and GPU—in order</p><p>to maximize the performance. Finally, we identify bottlenecks</p><p>and explain their impact on filter exchange latencies and update</p><p>rates.</p><p></p><p></p><p>Σε αντίστοιχα νέα έχουμε και την παρουσίαση της νέας σειρά DSP της Τ.Ι. TMS320C66x</p><p></p><p></p><p>CHIPMAKER Texas Instruments (TI) has released a digital signal processor (DSP) that it claims is the fastest on the market.</p><p>…</p><p>TI's multi-core DSP chips will be priced starting at $99 per 1,000 units and will be available for delivery in the first quarter of 2011. µ</p><p><a href="http://www.theinquirer.net/inquirer/news/1895018/texas-instruments-dsp-chip-claims-beat-gpgpus" target="_blank">http://www.theinquirer.net/inquirer/news/1895018/texas-instruments-dsp-chip-claims-beat-gpgpus</a></p><p></p><p>περισσότερα εδώ:</p><p><a href="http://focus.ti.com/dsp/docs/dspcontent.tsp?contentId=77428&DCMP=nysh_101109&HQS=Other+OT+c66multicore" target="_blank">http://focus.ti.com/dsp/docs/dspcontent.tsp?contentId=77428&DCMP=nysh_101109&HQS=Other+OT+c66multicore</a></p><p></p><p></p><p>Δεν έχω την ικανότητα / γνώσεις να παράξω αντίστοιχο κώδικα σε OpenCL δυστυχώς … αν και δοκίμασα το σχετικό api της ΑΤΙ … μου φάνηκε βουνό … </p><p>Ελπίζω σύντομα να δούμε και Open source υλοποιήσεις….</p></blockquote><p></p>
[QUOTE="tmjuju, post: 1417713, member: 2204"] FIR @ GPU Ένα από τα πρώτα Papers που βρίσκω που βασίζονται σε μοντέρνες υλοποιήσεις GPU. Εννοείται ότι η δουλειά του Koon ( [url]http://www.avclub.gr/forum/showthread.php?t=29204&page=2&highlight=koon[/url] ) στο diy δεν έχει θέση σε ένα ακαδημαϊκό paper :p , αλλά είμαι σίγουρος ότι σταδιακά θα δούμε και άλλα papers. HIGH-PERFORMANCE REAL-TIME FIR-FILTERING USING FAST CONVOLUTION ON GRAPHICS HARDWARE Institute of Technical Acoustics, RWTH Aachen University Aachen, Germany [email]fwe@akustik.rwth-aachen.de[/email] Institute of Technical Acoustics, RWTH Aachen University Aachen, Germany [url]http://dafx10.iem.at/proceedings/papers/WefersBerg_DAFx10_P76.pdf[/url] ABSTRACT In this paper we examine how graphic hardware can be used for real-time FIR filtering. We implement uniformly-partitioned fast convolution in the frequency-domain and evaluate its performance on a NVIDIA GTX 285 graphics card. Motivated by audio rendering for virtual reality, our focus lies on large-scale realtime filtering with a multitude of channels, long impulse responses and low latencies. Graphics hardware has already been used for audio signal processing — including FIR and IIR filtering with respect to offline and real-time processing. However, the combination of GPU computing and real-time conditions leads to a number of challenges that have not been reviewed in detail. The new contribution of this paper is an implementation and detailled analysis of a frequency-domain fast convolution method on GPUs. We discuss specific problems that emerge under real-time conditions. Our method allows to achieve an outstanding real-time filtering performance. In this work, we do not only regard a timeinvariant filtering, but also time-varying filtering, where filters are exchanged during runtime. Furthermore, we examine the opportunities of distributed computation—using CPU and GPU—in order to maximize the performance. Finally, we identify bottlenecks and explain their impact on filter exchange latencies and update rates. Σε αντίστοιχα νέα έχουμε και την παρουσίαση της νέας σειρά DSP της Τ.Ι. TMS320C66x CHIPMAKER Texas Instruments (TI) has released a digital signal processor (DSP) that it claims is the fastest on the market. … TI's multi-core DSP chips will be priced starting at $99 per 1,000 units and will be available for delivery in the first quarter of 2011. µ [url]http://www.theinquirer.net/inquirer/news/1895018/texas-instruments-dsp-chip-claims-beat-gpgpus[/url] περισσότερα εδώ: [url]http://focus.ti.com/dsp/docs/dspcontent.tsp?contentId=77428&DCMP=nysh_101109&HQS=Other+OT+c66multicore[/url] Δεν έχω την ικανότητα / γνώσεις να παράξω αντίστοιχο κώδικα σε OpenCL δυστυχώς … αν και δοκίμασα το σχετικό api της ΑΤΙ … μου φάνηκε βουνό … Ελπίζω σύντομα να δούμε και Open source υλοποιήσεις…. [/QUOTE]
Verification
Post reply
Home
Forums
Δικάναλος Ήχος
Do It Yourself
Διάφορες Kατασκευές
FIR @ GPU
Top
Bottom
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
Accept
Learn more…
