Hello folks Anyone know if this would be enough to decode oggs? AMD Elan SC520 133 MHz If not, what seems to be the lower limit required for decoding oggs? Many thanks, Kerry. <p>--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
Kerry (Kerry@email.message.co.nz) wrote:> Anyone know if this would be enough to decode oggs? > > AMD Elan SC520 133 MHzIt would probably be more accurate for you to attempt it on your own computer, and see whether it's up to the task, rather than asking a bunch of people who don't have your computer. :) I'm guessing it should be enough for realtime Vorbis decoding, if it has a floating point unit. -- Greg Wooledge | "Truth belongs to everybody." greg@wooledge.org | - The Red Hot Chili Peppers http://wooledge.org/~greg/ | -------------- next part -------------- A non-text attachment was scrubbed... Name: part Type: application/pgp-signature Size: 190 bytes Desc: not available Url : http://lists.xiph.org/pipermail/vorbis/attachments/20020627/50d4c52b/part-0001.pgp
The next up embedded chip i can find is the national semiconductor geode chips which i already know works. Then its onto the higher priced intel rage which usually still need heatsinks or transmeta which is overkill. That AMD chip seems very cheap and may work too... here tis a page: http://www.amd.com/epd/processors/4.32bitcont/14.lan5xxfam/24.lansc520/index.html and: http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/19751.pdf looks like it is two versions, a 486 and a Pentium. anyone tried a p133 to decode ogg files? <p> >>> baldusi@uol.com.ar 06/28/02 02:47 >>> Hello Greg, Thursday, June 27, 2002, 11:05:22 PM, you wrote: GW> Kerry (Kerry@email.message.co.nz) wrote: >> Anyone know if this would be enough to decode oggs? >> >> AMD Elan SC520 133 MHz GW> It would probably be more accurate for you to attempt it on your own GW> computer, and see whether it's up to the task, rather than asking a GW> bunch of people who don't have your computer. :) GW> I'm guessing it should be enough for realtime Vorbis decoding, if it GW> has a floating point unit. If I recall correctly that is an embedded AMD 486. <p><p>--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered. <p><p>--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
Kerry wrote:> > Hello folks > > Anyone know if this would be enough to decode oggs? > > AMD Elan SC520 133 MHz > > If not, what seems to be the lower limit required for decoding oggs?<p>Lucky you -- I'm trying to get an Elan 4xx (33 or so MHz non-fpu 486) to decode in realtime. I think the 520 will do fine, although the libvorbis implementation might be too slow. <p>Segher <p>--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
Kerry wrote:> Anyone know if this would be enough to decode oggs? > > AMD Elan SC520 133 MHz > > If not, what seems to be the lower limit required for decoding oggs?Some time ago I did some testings on an underclocked(!) Pentium classic @ 50 MHz (Win95, Winamp 2.7x, Peter's RC2 Plugin). This machine was able to decode ~96-112 kpbs RC2 files in realtime including streaming the PCM data to an old ISA MAD16 soundcard. All IDE-drives were in slow PIO mode. The Elan SC520 is based on AMD's 5x86, a 486 "on steroids". The 5x86 @ 133 MHz performs, roughly speaking, like a Pentium 75 on "real life" Windows-software (the 5x86 @ 133 MHz is therefore often called "5x86 PR75 133). It surely is able to to MP3 decoding in realtime: "This version plays a layer3,112Kbit/s,J-Stereo stream on my AMD 5x86/133Mhz with about 66% CPU load. (on the console, NO X11)" (taken from mpg123-0.59r README) Keep in mind that mpg123 is quite optimized. AFAIK Vorbis decoding still is a bit slower. Bit-rate "spikes" could be critical. The Elan SC520 has quite a small L1 Cache (16 KB) without a L2-Cache (I think there's no L2-cache controller in SC520). Of course, the integrated SDRAM-controller should be faster than those 486 FPM Chipsets - so the missing L2-Cache should not have a too big impact on memory-performance. Well, to be honest, I'm guessing here. By the way: The predecessor of NS Geode, the Cyrix MediaGX, also was based on a 486-like core (a Cyrix 5x86 - roughly the same performance as AMD 5x86). The MediaGX had integrated VGA (UMA), Sound, PCI etc... The very FPU-demanding Quake did run (roughly speaking) with Pentium 60 performance on a MediaGX @ 133 MHz. I do expect similiar results for the AMD 5x86/Elan SC520. <p><p>> Many thanks,> Kerry.bye, Maik Merten <p>--- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/ To unsubscribe from this list, send a message to 'vorbis-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.