2 MB /sec Network Speed | page 1 2
|
---|
|
---|
| | Gunnar von Boehn (Apollo Team Member) Posts 6214 12 May 2016 21:19
| Network download with 2 MB/sec
| |
| | Chris Marsden
Posts 36 12 May 2016 23:03
| Please tell us more!
| |
| | Simo Koivukoski (Apollo Team Member) Posts 601 13 May 2016 04:19
| Chris Marsden wrote:
| Please tell us more! |
It's download test from the local server: Vampire 600 V2 (core V3_3228_x12_127MB) AmigaOS 3.1 AmiTCP-3.0b2 D-Link PCMCIA cnet.device 1.9 ncftp-1.5.6 (HC_Reality.lha 55.9Mb / ~2148.85 kB/s)
| |
| | Chris Marsden
Posts 36 13 May 2016 07:02
| how much of an improvement over silver 5?
| |
| | Marcus Gerards
Posts 58 13 May 2016 09:30
| Yeah, it's stable now. :D Thanks Gunnar!
| |
| | Simo Koivukoski (Apollo Team Member) Posts 601 13 May 2016 15:37
| Chris Marsden wrote:
| how much of an improvement over silver 5?
|
Not in the speed but stability. If you didn't have any problems before you don't see any difference. This should solve freezing issue what some people experienced.
| |
| | Simo Koivukoski (Apollo Team Member) Posts 601 14 May 2016 05:43
| Vampire 600 V2 (core V3_3228_x12_127MB) AmigaOS 3.1 AmiTCP-3.0b2 plipbox (parallel port) v0.6 ncftp-1.5.6 (HC_Reality.lha 55.9Mb / ~136.01 kB/s)EXTERNAL LINK
| |
| | Mr-Z EdgeOfPanic
Posts 189 14 May 2016 10:03
| Simo Koivukoski wrote:
| Vampire 600 V2 (core V3_3228_x12_127MB) AmigaOS 3.1 AmiTCP-3.0b2 plipbox (parallel port) v0.6 ncftp-1.5.6 (HC_Reality.lha 55.9Mb / ~136.01 kB/s) EXTERNAL LINK |
Not bad at all, i expected a lot slower with plipbox.
| |
| | Manuel Jesus
Posts 155 28 May 2016 14:36
| Nice time to run some ethernet cable nthe basement Amiga lair. I was going to try wireless
| |
| | Simo Koivukoski (Apollo Team Member) Posts 601 25 Sep 2016 20:33
| Vampire 500 V2 (core V500_3478_x14_895) AmigaOS 3.1 AmiTCP-3.0b2 X-Surf-100 Zorro-II x-surf-100.device 1.11 ncftp-1.5.6 (HC_Reality.lha 55.9Mb / ~2171.46 kB/s)EXTERNAL LINK
| |
| | Simo Koivukoski (Apollo Team Member) Posts 601 25 Sep 2016 20:37
| a quick comparison chart:Vampire 500 V2 (core V500_3478_x14_895) X-Surf-100 Zorro-II ~2171.46 kB/s Vampire 600 V2 (core V3_3228_x12_127MB) D-Link PCMCIA ~2148.85 kB/s Vampire 600 V2 (core V3_3228_x12_127MB) plipbox (parallel port) v0.6 ~136.01 kB/s
| |
| | John Heritage
Posts 111 26 Sep 2016 03:31
| That's pretty decent! Just curious - what do you think is holding it back from maxing out 100 mbps (~ 10 or 12.5MB/sec)? Would RTG speed this up? Maybe increasing the TCP/IP packet size? The FTP software being inefficient? I'd think the PCMCIA bus (16-bit * 7.14 or 8 mhz) should be fast enough - it's not DMA, but that equates to ~14MB/sec theoretical max. The Apollo core is probably equal to a Pentium-133 or 166 in integer performance? should be enough to max out a 100 mbps / 10MB/sec transfer.
| |
| | Gunnar von Boehn (Apollo Team Member) Posts 6214 26 Sep 2016 05:42
| John Heritage wrote:
| I'd think the PCMCIA bus (16-bit * 7.14 or 8 mhz) should be fast enough - it's not DMA, but that equates to ~14MB/sec theoretical max.
|
The AMIGA BUS runs at 7 Mhz. But one bus-access uses a 4 cycle protocol. This means the max speed possible is 2 Byte *7 /4 = 3.5 MB
| |
| | Gregthe Canuck
Posts 274 26 Sep 2016 08:23
| Would be interesting to see if Roadshow 68K was any faster.
| |
| | Marcus Gerards
Posts 58 26 Sep 2016 13:29
| gregthe canuck wrote:
| Would be interesting to see if Roadshow 68K was any faster.
|
Roadshow IIRC has several features ensuring transfer quality which AmiTCP does not have. This results in AMiTCP being slightly faster in some tests (not significantly). I prefer Roadshow anyway. More interesting would be the performance of the X-Surf 100 in Z3 mode. We'd need a Vampire 3000/4000, of course. Another revision of the V500/V600 with an integrated NIC should even top such a Zorro3-configuration.
| |
| | Gregthe Canuck
Posts 274 26 Sep 2016 15:12
| The nice thing with Roadshow is that this is a supported product. Given enough incentive I'm sure the developer could be convinced to optimize for the Apollo core. :)
| |
| | Szyk Cech
Posts 191 26 Sep 2016 15:21
| Shut up!!! Don't even mention about V4000!!! They said it is very unlike to arise. I have question for you (V4000 fan boyz): Do you prefer stand alone Vampire, or V4000?!? I think many generations Amiga fans wait for former, and very little wait for the later. And more: You have about half Apollo performance in old 68060 turbo cards... So you are not in so desperate situation as A500, A600 users. As you know 68060 turbo cards exists even for A1200, so it is unknown for me why they want to do V1200. Maybe this is substantial market. I can understand that. How ever problems with buying edge connectors for V1200 makes this project very unlikely to be done. Maybe that will be better to concentrate every further efforts for stand alone...
| |
| | John Heritage
Posts 111 26 Sep 2016 15:23
| Gunnar von Boehn wrote:
| The AMIGA BUS runs at 7 Mhz. But one bus-access uses a 4 cycle protocol. This means the max speed possible is 2 Byte *7 /4 = 3.5 MB
|
Thanks Gunnar! I assume there are no tricks to reduce this cycle count? Like blanking the screen on the Atari 8bit would do to free up some chipset cycles? (Apollo showing >50% of that 3.5MB/sec for netwrok transfer tells me the CPU is maxing out the bus totally, which is impressive).
| |
| | Marlon Beijer
Posts 182 26 Sep 2016 15:24
| Szyk Cech wrote:
| 68060 turbo cards exists even for A1200, so it is unknown for me why they want to do V1200 |
They cost a fortune, not something everyone can afford.
| |
| | Simo Koivukoski (Apollo Team Member) Posts 601 26 Sep 2016 15:35
| NETIO Benchmark result with X-Surf-100 Zorro-II and Vampire 500 V2:New Shell process 6 6.Ram Disk:> netio -p 18000 -t 192.168.2.1NETIO - Network Throughput Benchmark, Version 1.32 (C) 1997-2012 Kai Uwe Rommel TCP connection established. Packet size 1k bytes: 1792.51 KByte/s Tx, 2013.84 KByte/s Rx. Packet size 2k bytes: 1925.54 KByte/s Tx, 2146.03 KByte/s Rx. Packet size 4k bytes: 2072.29 KByte/s Tx, 2199.50 KByte/s Rx. Packet size 8k bytes: 2131.46 KByte/s Tx, 2232.50 KByte/s Rx. Packet size 16k bytes: 2148.60 KByte/s Tx, 2247.70 KByte/s Rx. Packet size 32k bytes: 2111.17 KByte/s Tx, 2243.38 KByte/s Rx. Done. 6.Ram Disk:>
| |
|