Has anyone tweaked there Roadshowcontrol settings to run more efficently/quickly with a high speed ADSL connection.
I know there is information already out there concerning this but I'm just pondering if the SAM ethernet ports are different for example.
Ranger also reports MTU and Hardware is at 1518 I decided to take off 40 from that and set that as my MTU in Roadshow.
I think it would have been nice to have a toggle, either set it to something or 0 to disable and then the hardware figures it out. Once established it also changed relevant buffers and caches.
Before I saw the Ranger information I had it set at 1500 which is the setting at my provider. I didn't take off 40, that's just a new thing and seems to be related to other systems, must be a reason for it though.
Not at home, at work with this horrible IBM LOL so can't provide any cut and paste
Thanks for any information
~Yes I am a Kiwi, No, I did not appear as an extra in 'Lord of the Rings'~ 1x AmigaOne X5000 2.0GHz 2gM RadeonR9280X AOS4.x 3x AmigaOne X1000 1.8GHz 2gM RadeonHD7970 AOS4.x
Slayer wrote: Has anyone tweaked there Roadshowcontrol settings to run more efficently/quickly with a high speed ADSL connection.
I know there is information already out there concerning this but I'm just pondering if the SAM ethernet ports are different for example.
Ethernet is a standardized technology, with standardization covering both the electrical aspects and the makeup of the data that is committed to the wire. There are differences between the electrical and physical dimensions (connector types, cabling, etc.), but what matters for ADSL isn't this, but how the data on the wire looks like. And for consumer use, that's very narrowly defined: 1500 bytes payload, with a 14 byte header that tells every listener where the data comes from, where it goes to, and what protocol it uses. This is what counts and the SAM Ethernet hardware doesn't do this any differently.
Quote:
Ranger also reports MTU and Hardware is at 1518 I decided to take off 40 from that and set that as my MTU in Roadshow.
What counts is the payload size Ethernet permits, and that's 1500 bytes. You should use this lower number as the basis for your calculations.
Quote:
I think it would have been nice to have a toggle, either set it to something or 0 to disable and then the hardware figures it out. Once established it also changed relevant buffers and caches.
Before I saw the Ranger information I had it set at 1500 which is the setting at my provider. I didn't take off 40, that's just a new thing and seems to be related to other systems, must be a reason for it though.
Not at home, at work with this horrible IBM LOL so can't provide any cut and paste
Thanks for any information
Tweaking the MTU value doesn't necessarily result in noticeable improvements in terms of performance, unless you are connecting directly to the Internet through PPPoE, which you really should not (use a gateway router instead, please: it's safer and simplifies a whole lot of things).
Some useful tweaks involve internal configuration parameters of the TCP/IP stacks related to buffer sizes and packet assembly. Try entering this in the shell:
roadshowcontrol save set tcp.do_timestamps 0 roadshowcontrol save set tcp.do_win_scale 0 roadshowcontrol save set tcp.mssdflt 1460 roadshowcontrol save set tcp.recvspace 65536 roadshowcontrol save set tcp.use_mssdflt_for_remote 0
Thanks for the indepth overview and those settings are a good reference, I think mine reflect that or close too it, I'll check I have them pasted in another thread on the board.
Thanks! I'm sure others will also benefit
~Yes I am a Kiwi, No, I did not appear as an extra in 'Lord of the Rings'~ 1x AmigaOne X5000 2.0GHz 2gM RadeonR9280X AOS4.x 3x AmigaOne X1000 1.8GHz 2gM RadeonHD7970 AOS4.x
Are the above settings the best we can have with Roadshow?
I just finished to install linux on my Sam and there is a huge difference in download speed.
I tried a website wich offers a speed test and I get 1.244 kbit/s on OS4.1u4 with OWB and more than 10times that value on the very same Sam using Debian (first lenny, now squeez) !! :-O (15.876 kbit/s)
The results are almost the same on AmigaONE-XE which is even slower than sam (1.211 kbit/s)
any ideas?
Results are not changing if I use DHCP against Fixed IP
@Amigo1 That does seem slow. I test download speed at Testmy.net and my SAM Flex gets 2.3 Mbps with OWB but almost 4 Mbps with MUI-OWB. Apparently software and settings make a difference. I'm on a cable connection which can supposedly get 10 Mbps but I never get that.
Amiga X1000 with 2GB memory & OS 4.1FE + Radeon HD 5450
Not sure what the need is to go poking around in the tcp/ip stack. Usually just plug in the ethernet cable and forget about it. At least thats what I did on my SAM and I've speeds of 1MB a second - same speeds as my mac and win7 netbook.
I used to use an MTU size of 576 and 1088 between MiamiDx on AmigaOS3.9BB2 and Linux and had a throughput from Samba on the A4000 serving over 10Mbit Ethernet of 100KBytes/second transfer rates...
Literally fast enough to watch video stored on the Amiga from the Linux machine (I was also viewing from windows)
but if you really want to tweak for speed then you need to consider ALL the devices on the network concerned and what timings they all have.
xenic wrote: @Amigo1 That does seem slow. I test download speed at Testmy.net and my SAM Flex gets 2.3 Mbps with OWB but almost 4 Mbps with MUI-OWB. Apparently software and settings make a difference. I'm on a cable connection which can supposedly get 10 Mbps but I never get that.
I just tried your website, I get 987 kbps.. even less than where I tried. Debian downloads at 16.419 kbps. That's odd
djrikki wrote: Not sure what the need is to go poking around in the tcp/ip stack. Usually just plug in the ethernet cable and forget about it. At least thats what I did on my SAM and I've speeds of 1MB a second - same speeds as my mac and win7 netbook.
yes, I would do the same, but since I got my new cable connection and the shown speed is 1/30 of what it should be, while Debian on the same machine get's at least half of the should-be speed (note I get 30.144 kbps on my mac), I'd like to figure out the reasons.
The default Roadshow settings got me the first batch so slow results, then tried the settings posted above, without noticeable difference. So I'd like to poke around some more, or at least be able to set back to default values, wich -silly me- I did not write down prior change.
Ranger also reports MTU and Hardware is at 1518 I decided to take off 40 from that and set that as my MTU in Roadshow.
How do you set MTU ? RoadShowControl does not mention MTU?
I did the speedtest and for my SAM440ep (which happens to have the same settings as suggested in htis thread by Olsen) i have a download speed of 2.1 MB/s
Still far away from the nearly 16 MB/s reported by Amigo1 with linux.
@JosDuchIt I think were getting the speed units confused. I think Amigo1 reported 16kbps which is rediculously low but would be unbelievabley high if it were 16Mbps. Your speed is in different units that I gave: MBps = megabytes per second Mbps = megabits per second kbps = kilobits per second There is a big difference between MBps and Mbps. If you actually meant Mbps then your speed is equivalent to mine for a standard cable connection. I use Testmy.net because I had my sister run that speed test on a cable connection in a location 100 miles from me but still using Comcast cable. She got the same ballpark result as I did: between 2.1 Mbps and 2.4 Mbps.
I don't know how reliable these speed tests are. For one thing, I can't even use most of them because they require Flash for the test. Other sites just give me rediculous results. For example, whatismyip.com gives me 0.82 Mbps download and 2.51 Mbps upload with OWB but 1.3 Mbps download and 16 Mbps upload with MUI-OWB. I don't know of any ISP that gives you 3-10 times the upload speed as download speed.
I ran a "realworld" download test by downloading the 52MB amicygnix-sdk-lite.lha file from OS4 depot with OWB and it took 4 minutes 45 seconds. That calculates out to approximately 1.4 Mbps which is reasonable. I ran the same test with MUI-OWB and the download took 1 minute 55 seconds. That calculates out to approximately 3.84 Mbps which is great! It appears that browser software can make a huge difference in download speed results so if anyone else wants to post results they should mention which browser was used for the test.
Amiga X1000 with 2GB memory & OS 4.1FE + Radeon HD 5450
Ranger also reports MTU and Hardware is at 1518 I decided to take off 40 from that and set that as my MTU in Roadshow.
How do you set MTU ? RoadShowControl does not mention MTU?
I did the speedtest and for my SAM440ep (which happens to have the same settings as suggested in htis thread by Olsen) i have a download speed of 2.1 MB/s
Still far away from the nearly 16 MB/s reported by Amigo1 with linux.
Actually I wrote 16.419 kbps which translates roughly to 2 MiB/sec. But that is using Linux Debian. On OS4.1u4 I get approx 1.204 kbit/sec or 150 kiB :-/
You were almost right, as I posted above I reported a value ranging from 15000 to 16000 kbps (15 to 16 Mbps) with Linux (and sadly much less with OS4)
I have a cable connection and at the moment it seems to work pretty well. on macos I get values ranging from 25000 to almost 32000kbps (32Mbps) out of 34000
I intentionally obmitted the "." and/or "," to avoid confusion because of the inverted meaning of those mathematical symbols here and overseeas.
Noticed pretty high CPU load with low speed results on SAM440ep, when doing those tests in OWB. PPC440 is a telecommunication chip, so it should be able to handle TCI/IP stack without problems, so, browsers are the bottlenec?
(I get below 1Mbit test results with SAM, when the connection can handle 10-18Mbit when tested with x86 Linux system.)
On the other hand ... with the mentioned test file (amicygnix-sdk-lite.lha) I get 400k...900k bytes / second with IBrowse and 1200...1500k bytes with firefox on x86. So, with IB and real word test, the speed does not seem THAT bad, even though it could be a higher.
- Kimmo --------------------------PowerPC-Advantage------------------------ "PowerPC Operating Systems can use a microkernel architecture with all it�s advantages yet without the cost of slow context switches." - N. Blachford
On Sam440ep (o/c at 733Mhz), I can get speeds as up as 7MB/s (nearly the max the ethernet chip can achieve) under AmigaOS 4.1 Update 4 (this used to be the case formerly by the way).
On AmigaOne, with the integrated chipset, I can get speeds as up as 3MB/s.
To really test your speed, try to download multiple files with OWB from a really fast server.
Use c:SampleNetSpeed to monitor your connection speed.
Bear in mind that the higher the speed you get and the higher your CPU will be used.
I'm using a Cable Connection from the provider Numericable (100Mbs DL / 5Mbs UL).
K-L wrote: On Sam440ep (o/c at 733Mhz), I can get speeds as up as 7MB/s (nearly the max the ethernet chip can achieve) under AmigaOS 4.1 Update 4 (this used to be the case formerly by the way).
On AmigaOne, with the integrated chipset, I can get speeds as up as 3MB/s.
To really test your speed, try to download multiple files with OWB from a really fast server.
Use c:SampleNetSpeed to monitor your connection speed.
Bear in mind that the higher the speed you get and the higher your CPU will be used.
I'm using a Cable Connection from the provider Numericable (100Mbs DL / 5Mbs UL).
That really sounds good. 4MB/s would make me happy. Would you mind posting the settings to achieve that?
btw, I just re-tried, I can only confirm the low speed.
Tried to download libopenssl. from os4depot. it takes 1minute 12 seconds on Sam (800MHz) and 1 minute 13 seconds on A1XE.
Sam Debian 8 seconds MacOS X MPB 4 seconds.
All on the same router, Static Address assigned to all of them, but tried with DHCP and interchanging the ethernet port the computers are connected to the router. No noticeable difference.