Jump to content


Photo

Long-run USB alternative idea - router

  • Please log in to reply
54 replies to this topic

#26 ccs_hello

ccs_hello

    Fly Me to the Moon

  • *****
  • Posts: 6548
  • Joined: 03 Jul 2004

Posted 08 January 2013 - 06:55 PM

... Orion Star Shoot 3 mono / QHY ...

At least I know these two are using the "double boot" method which is tough to emulate. Many astroimager use streaming in high-speed isochronous transfer mode which is also hard.

Elite Silcon (E3868)'s method just need one box on the USB device end due to its "USB over IP" strategy, while icron needs two (one on device end and the other on PC end) which is using "USB over proprietary" method.

In both cases, there is a short cut: bypassing two-way handashake (which is the reason why the USB max. distance is limited) and just "transmit in blind".
In plain English, say a supplier kept sending bags of concrete to a build site. At some point, the project manager in the build site will have to say, stop (or slow down) I am too busy here. Or he might have to tell the sender that a specific bag has defect material and please resend that one. Regular USB with two-way handshake can do that. The work around method simply doesn't care and claims the world peace achieved.
Will latter work all the time? Luck will tell :).

P.S. recent refurb Core 2 Duo laptops (many formerly business grade) can be obtained for less than $200, Win XP Pro included.

Clear Skies!

ccs_hello

#27 microstar

microstar

    Apollo

  • -----
  • Posts: 1231
  • Joined: 05 Jan 2008
  • Loc: Canada

Posted 08 January 2013 - 08:02 PM

As I said, the ICRON Ranger 2104 (now replaced by the 2204) USB over Cat5 has no problem with collecting data from a dual imaging rig with an Atik 383L+ mono CCD and Canon XSi while at the same time guiding with a QHY5 webcam and PHD and controlling the mount, focuser and sky monitor. I know when I told the QHY5 distributor here in Canada that I was doing this over a single Cat5 cable he was surprised. It costs a little more (but look around, I saw one Canadian distributor clearing them out at a discount) but it works with high-speed isochronous transfer and multiple devices simultaneously. Although some things didn't like being connected via the hub attached to the 2104 and needed to be connected directly to the unit, other than having to play with the arrangement it has worked pretty flawlessly for me.
...Keith

#28 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 08 January 2013 - 09:46 PM

Oh, I didn't say that this was with 100' of Cat6 cable as well.

My only issue with all of these devices is the lack of troubleshooting capability at the telescope. For instance, some times you want to be right there to test things or figure out why your filter wheel/focuser is on the fritz. Using one of these requires you to run back n forth or utilize a tablet...

I'll stick to a cheap computer at the 'scope.

#29 Charlie B

Charlie B

    Viking 1

  • -----
  • Posts: 967
  • Joined: 22 Mar 2008
  • Loc: Sterling, Virginia

Posted 09 January 2013 - 05:31 PM

such as leaving a laptop there and VNC'ing into it from the desktop


This is the method I use. I got a new laptop for Christmas and now my old one is available for remote control. I use UltraVNC and it works perfectly through my wireless router. I've connected up to 300 feet with good signal.

I still use a USB hub to the remote computer, but only need a single cable. The hub has the advantage of being able to easily route camera/filter/focuser cables close to the cg of the mount with not a lot of long cables to the computer. I could even mount the hub on the ota.

Regards,

Charlie B
Regards,

Charlie B

#30 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 09 January 2013 - 07:30 PM

Part 2 /Day 2 continuation of evaluation of remote control/data transfer for imaging...

To see how the results of the speed tests using the USBgear 4 USB port ethernet hub over a 100 ft Cat-5E cable that I reported in my post above compared to the USB/WiFi 4 port hub and to the Tether Pro Active USB cables that I have been using very satisfactorily for the past 3 years, I accomplished several more tests this afternoon.

The conditions and parameters for these tests are the same as for the tests yesterday, the same HP DV6T laptop was used with the same HDspeed test software, only the communications protocol was changed along with the necessary communications devices.

The first test this afternoon was to see how well the IOgear 4 port USB over WiFi hub performed...

The communications path is from a wireless-N adapter built into the HP Laptop to a Verizon Jetpack 4G/Wireless-N router (part of my inhouse wireless network) to the IOgear USB/WiFi 4 port hub. Distances = 4 ft from Laptop to WiFi router, 8 feet from WiFi router to USB/WiFi 4 port hub so RF signal levels were not a limiting factor.

Using the same Passport HD USB3.0 1TB drive with the 3 ft USB3.0 to USB2.0 adapter cable, READ tests of 3 minute elapsed time resulted in 64 Kb block size being optimum and best speed reached was 1.1M bytes/sec.

I then decided to make a comparison of the tests results run so far to the Tether Pro Active USB extension cables tested in the same configuration as I use them when imaging...

My first test of a Tether Pro cable was to attempt to access the Passport USB drive plugged into a 15 meter Tether Pro cable plugged directly into a USB2.0 port on the laptop. No hub, just the drive on one end of the cable and the other cable end plugged directly into the Laptop.

Nothing, the light on the drive came on but the computer refused to recognize the drive...

That had me puzzled for a moment, then I realized that the Passport being a 1TB USB3.o drive probably draws more current than the extension cable being USB2.0 can deliver considering the length of the cable has to also power the repeater modules. At that point I put the USBgear 4 port USB2.0 externally powered hub on the far end of the Tether Pro USB extension cable and plugged the Passport drive into the hub. ((Note that the USBgear 4 port hub also has a USB link port making it usable as either TCPIP or USB linked hub).

Instant success... And not surprisingly, the HDspeed test utility showed nearly the same rate as the Passport drive plugged directly into the laptop with the 3 ft cable. Read time for elapsed 3 minute test over the 15 meter Tether Pro cable was 25.4M byte/sec.

At that point I added another 7.75 meter Tether Pro cable in series with the 15 meter cable for a total of 22.75 meters and speed dropped very slightly to 25.1M bytes/sec.

I decided to see what would happen if I added my other 7.75 Meter Tether Pro USB cable for a total of 30.5 meters which exceeds the allowable useable length according to the Tether Pro documentation which states the usable limit is 25 meters... Windows immediately produced an error message stating number of USB repeaters = 5 exceeds allowable limit. (apparently the hub is counted as a repeater as well)...

So the concept of adding more Active USB repeater cables even if reduced performance is acceptable is not possible... It would seem the 25 meter limit as stated is the functional limit with no exception...

After two days of testing various link communications methods and protocols, the conclusion is:

Passport 1TB USB HD plugged into Laptop = 26.1M bytes/sec.

25 meters or less, Active USB cables = 25M-26M bytes/sec.

Over 25 Meters with USB/IP hub = 14M-15M bytes/sec.

WiFi - (not distance tested) = 1.1M bytes/second.

Note, the actual speeds reported may vary with equipment used. The importance here is that for all tests I have run I have used the same Laptop and the same USB HD and the same HDspeed test software to keep the results standardized. The speeds as I report are in no way indicative of actual product specifications, this is purely a series of comparison tests...

That is the end of these tests as far as I am concerned, my conclusion is that WiFi is not suitable for combined mount control, imaging data flow, and Guiding camera data all combined on one WiFi link using a 4 port USB2.0/WiFi hub.

As to USB extension cables vs USB/IP, more testing when the weather improves to see how each protocol performs dynamically with full imaging operation and/or need for length beyond the 25 meter limit of USB will be the deciding factor...

#31 A. Viegas

A. Viegas

    Apollo

  • *****
  • Posts: 1085
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 09 January 2013 - 10:44 PM

great work Mitch. Looks like a string of USB repeater cables is hard to beat unless you need to go much more than 100ft. (especially when you consider cost!).

Al

#32 ccs_hello

ccs_hello

    Fly Me to the Moon

  • *****
  • Posts: 6548
  • Joined: 03 Jul 2004

Posted 09 January 2013 - 10:53 PM

USB repeater by the book method, see this CN thread.

Clear Skies!

ccs_hello

#33 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 09 January 2013 - 11:56 PM

great work Mitch. Looks like a string of USB repeater cables is hard to beat unless you need to go much more than 100ft. (especially when you consider cost!).

Al



Hi Al,

Yes, these tests turned out to surprise me more than I was expecting. I had not expected the Tether Pro cables to perform as well as they do, I was expecting more loss at 15 - 25 meters than I found with them. They are virtually lossless for all practical purposes...

And I was expecting the TCPIP link to blow the sox of the USB cables as far as speed is concerned which it did not, another disappointing result...

However, I do think the TCPIP performed well considering there was very little change by adding a 100 ft. cable. In hind sight, thinking about this TCPIP Ethernet link test afterwards, there is one more test I should run, and that is to link another 100 ft cable making the length 200 ft...

I have several of the 100 ft Cat-5e made up cables and an additional 50 ft. cable. I have three RJ-45 couplers so it would be a simple test to expand the results to include test results with a longer run of Cat-5e cable...

However, in my own setup, I do not need more than 100 ft. maximum. I'm working with 15 meter USB cables which are plenty long enough... So, whether I actually will be inspired to run several hundred feet of cable out is an unknown, I'll have to think about it for a bit... I have yet to re-coil the 100 ft Cat-5e cable I have strung around the place now, so maybe tomorrow I'll get inspired...

There was another comparison result that came about in this testing that I did not bother to write up. I have a IOgear 4 port USB/IP hub, a retail consumer grade loaned on trial, and I did not consider it to be a contender. As like most IOgear, its a really cheap design and the specifications are very vague however its labeled and sold as a USB2.0 complaint device. Therefore I did plug it in and configure it for the network however the most I could get out of it was 3.7M bytes/sec. using a 10 ft. CaT-5 cable...

Fortunately, I don't own this one, its going back where it came from as soon as I can get it packaged up and shipped out...

Just the same, it was interesting to have a second USB/IP hub here to compare to the USBgear USB/IP unit...

The biggest disappointment was the USB over WiFi test although I was expecting that result, just not as bad speed wise as it turned out to be, I'd hoped to get somewhere about 4M - 5M bytes/sec. however again, the manufacturers specifications were misleading, it turned out to be a 802.3g device, not wireles-N, the difference being 802.3g is a single band simplex protocol as compared to the current technology wireless-N being multiband full duplex capable...

So, if I can find the right gear, the USB over WiFi test may be retested when time allows... Probably a project for next winter...

#34 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 10 January 2013 - 01:26 AM

Well, a quick answer to USB over TCPIP on Cat-5E cables longer than 100 feet, a quick test plugging two 100 Ft. cables together doesn't work... It looks to be too much length for the 10-100-1000 Gigbit network. The Laptop knows something is plugged in but is unable to communicate. Looks like the Ethernet router Switch indicator lights are showing a 10base-T speed connection with 200 ft of cable where as with a single 100 ft. Cat-5 cable, it indicates high speed with both indicators lit... This could be the result of a poor quality or faulty cable as one of the two is one I've had for a while and all other tests were done with a brand new cable...

However, using a 6 ft Cat-5e patch cable to avoid unrolling the older 100ft. Cat-5e cable allows a test of the second cable and the coupler plugged into the Ethernet switch to Laptop and the coupler old cable combination tests OK...

Apparently the limiting factor is network speed/signal strength over more than 100 ft. of Cat-5E cable. As what I am using is inexpensive cable ($12.95/100 ft.) purchased on Amazon.com, possibly there is a better quality of network cable that might extend the range...

#35 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 13 January 2013 - 05:40 PM

Over the weekend, thanks to foggy/rainy weather, I had time for follow up testing in my process of deciding what is the most efficient protocol and equipment for remote control/data collection of my imaging setup.

In spite of the initial results showing that Tether Pro Active USB extension cables at 25M bytes/second at maximum tested length of 25 meters being the clear winner, my second choice being a USBgear 4 port USB/IP hub being the second most efficient method at 15M bytes/second over 110 feet of Cat-5e cable (100 ft to router plus 10 ft to the laptop from router), I decided the next test would be to see how well the USBgear hub over the 100 ft Cat 5 cable would be affected by multiple active Devices...

I also wanted to evaluate which of the two cameras would load the network the heaviest, the Canon 60Da or the Orion SSAG guiding camera. This is a test prompted by a post by Guylain Rochon that he has found the SSAG camera to be the heaviest load on communications in his setup.

To this end I plugged the USBgear 4 port USB/IP hub into my Ethernet router, plugged the Canon 60Da, the Orion SSAG autoguider, and the Passport HD into the USB hub.

With all three USB devices active, I started PHD, connected the autoguider camera pointed at a bright light to saturate the sensor, and started BYEOS live view connected to the Canon 60Da. I then started HDspeed test to monitor efficiency/speed of 256Kb block size continuous reads of the HD as an indicator of network load...

Interestingly, the combined effect of all three USB devices in operation simultaneously showed about a 10% drop in read speed of the HD.

To determine which USB device was the cause of the heaviest network load, I alternately disconnected the Orion SSAG and the Canon 60Da and while each was operating independently, tried various exposure rates.

The end result was as Guylain reports, the Orion SSAG caused the read rate of the HD to drop about 10% while the 60Da in either live view or imaging mode caused only a slight drop of a percent or two worse case. To my surprise, the 60Da active in Live View mode had little effect as far as network loading as indicated by read speed of the HD...

My conclusion is that if I continue to use USB active extension cables in the future, I will plug the SSAG camera into the dedicated cable and the 60Da into the USB hub which is the reverse of my previous setups.

I'm also reasonably convinced that USB over the TCPIP Ethernet link is adequate to handle the multiple USB inputs, that the 10% reduction in speed is virtually transparent to the practical operation of the cameras based by observing the live view of the Canon 60Da while the SSAG is running in saturated mode at various exposure rates also tested by making 5 test exposure with BYEOS in imaging mode transferring 20M RAW image files to the laptop over the link. A few instantaneous speed reduction spikes were observed in the HDspeed test graph however the reported speed average did not change...

Moving on...

After a night to think about the poor performance of the USB/WiFi link at 1.1M bytes/sec., as I am still very much interested in this method, I decided another test was worth conducting. This test to be using a WiFi link between the HP DV6T laptop with internal Wireless-N adapter using protocol to my inhouse i7 CPU quipped PC which has a Intel Wireless-N network adapter installed. The connection is through the Verizon 4G/Wireless-N router which is how my inhouse wireless network normally routes between the Laptop and the PC.

However, what I found was that the HDspeed test utility will not recognize the USB Passport HD over the link through the second PC via the wireless link. I believe this is because a USB hub uses USB repeater protocols and a PC does not, therefore the HSspeed test utility can not see the Passport HD through the PC as it does when using the USBgear USB/IP hub...

However, as one phase of my original testing of the 4 port USB/WiFi hub was a visible test of the 60Da in live view mode, the result being a lot of delay and dropped frames, I decided to test the WiFi link between the two computers using Win 7 remote desktop function and repeating the 60Ds Live View mode. Again, there was an increased visible delay but no obvious dropped frames that I could see... Using a true Wireless-N WiFi link does seem to demonstrate improved performance over the original WiFi test however is still noticeably slower than either the USB/IP Ethernet link or the use of Active USB extension cables.

My interpretation of the result of this crude live view comparison test using Win 7 RD shows that the link between the two PC's being a true wireless-N format as compared to the 4 port USB/WiFi hub being Wireless-G protocol hub. However, simply by switching between the two PC's shows the WiFi Wireless-N link is still considerably slower than the TCPIP Ethernet link...

AT this point, I am satisfied that the Tether Pro Active USB cables will continued to be used in my remote installation imaging setup however I also plan to continue testing and evaluation of the USB over TCPIP link as I believe it has adequate efficiency/speed and my guess at the moment has enough benefits with adequate performance to equal the active USB extension cable when in actual use when imaging... In part this is that I believe when used for actual imaging, the load on the link is not as continuous or as heavy as what I have generated for testing purposes...

#36 A. Viegas

A. Viegas

    Apollo

  • *****
  • Posts: 1085
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 13 January 2013 - 06:05 PM

Mitch, thanks for the analysis. Certainly suggests to me that using a long USB repeater chain is the preferred method. One question I still have, I think it was ccs_hello who said that there was a limit on the number of USB repeater hubs you can have. Is this a limit just for XP? Can windows 7 handle more than that? I also think he said a 7 port hub counts as two repeaters? So how can you get to 25m using the tether pro cables? That would suggest 5 5m repeaters plus your hub... Hopefully the answer is that there is in fact no limit on e USB repeater chain?

I also have a 100ft cat5 and USB dongle set, but I only use that for low speed devices like GPUSB ST4 box and USB to serial for planetarium or nexRemote. Interesting I tested plugging in is 100ft USB 1.1 into my USB 2.0 hub and testing the speed on three 5m repeater cables. Funny enough I clocked consistent 992kbps on the USB 1.1 irrespective of where it was plugged in. However the USB 2.0 repeater cables averaged 19Mbps to 23Mbps depending on which port I used on e hub. When I did not use the slower USB 1.1 plugged into the same hub, I achieved closer to 24Mbps, but only on one of the 4port hub connectors.

Net, net my setup is probably just going to continue as it... I ordered another 2 5M repeater hub cables, so I will have 5 segments and two hubs, one hub after 2 segments and another hub at e end of segment 5. I will drive two telescopes and 3 cameras. Let's see how it goes,

Al

#37 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 13 January 2013 - 06:54 PM

Mitch, thanks for the analysis. Certainly suggests to me that using a long USB repeater chain is the preferred method. One question I still have, I think it was ccs_hello who said that there was a limit on the number of USB repeater hubs you can have. Is this a limit just for XP? Can windows 7 handle more than that? I also think he said a 7 port hub counts as two repeaters? So how can you get to 25m using the tether pro cables? That would suggest 5 5m repeaters plus your hub... Hopefully the answer is that there is in fact no limit on e USB repeater chain?

I also have a 100ft cat5 and USB dongle set, but I only use that for low speed devices like GPUSB ST4 box and USB to serial for planetarium or nexRemote. Interesting I tested plugging in is 100ft USB 1.1 into my USB 2.0 hub and testing the speed on three 5m repeater cables. Funny enough I clocked consistent 992kbps on the USB 1.1 irrespective of where it was plugged in. However the USB 2.0 repeater cables averaged 19Mbps to 23Mbps depending on which port I used on e hub. When I did not use the slower USB 1.1 plugged into the same hub, I achieved closer to 24Mbps, but only on one of the 4port hub connectors.

Net, net my setup is probably just going to continue as it... I ordered another 2 5M repeater hub cables, so I will have 5 segments and two hubs, one hub after 2 segments and another hub at e end of segment 5. I will drive two telescopes and 3 cameras. Let's see how it goes,

Al


Hi Al,

The limit of repeaters in linked USB cables is a function of the USB protocol I believe, it is not dependent on the Windows operating system as far as I know. I have run into that here with the Windows 7 Pro 64 bit operating system on both of my computers. I believe I reported on that in one of my previous posts.

However in my testing, I have used a passive 10 ft USB extension cable plus a 9 ft USB device cable on the far end of daisy chained Active USB cables to the hub meaning I was testing with 25 meters of cable with 3 repeaters plus 6 meters of passive cable with a 4 port USB hub at 31 meters total. This seems to work as long as the active USB cables are closest to the computer. It would seem, at least in my testing, that a total of 4 repeater IC's is the maximum allowed.

As to a 8 port hub having two repeaters internally, I can not confirm this, I do have a 7 port IOgear USB hub that has proven not to work correctly however I have never seen the error message when using it that it has exceeded the allowable number of repeaters. My opinion is that the 7 port USB HUB is simply another piece of IOgear cheap junk... I've been burnt 3 times with IOgear junk, I will never purchase an IOgear product again...

BTW, for what it is worth, in my previous report/posts, I referred to using 7.75 meter and 15 Meter Tether Pro Cables, that was an error on my part, the 7.75 meter cables are actually 9.75 meter cables... I happened to be looking on the Tether Tools site this afternoon and realized my mistake.
http://www.shop.teth...-Cable-16-32...

As to daisy chaining 5 meter USB cables, I think you will be limited to 3 active cables plus a passive cable, and a hub maximum based on my experiments however i do not know anything about what cables you are using...

What I post about is based on what I have done myself and know works which means I could be wrong about what you discover in your setup...

#38 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 13 January 2013 - 07:26 PM

Al,

I meant to comment in my previous post, I have a couple sets of the USB to Cat-5 adapter dongles on 100 ft Cat 5 cables and they do work however at slower speed than the Active USB extension cables.

I have not attempted to do a benchmark speed test on them as I've had them for a while and since switching to the active USB extension cables, I've simply not had a need to use them...

That would be a simple test to run, You've piqued my curiosity, think I'll give that a try just to see where they fit in the scheme of link efficiency...

#39 A. Viegas

A. Viegas

    Apollo

  • *****
  • Posts: 1085
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 13 January 2013 - 08:11 PM

Mitch,

I currently use 2 ├╝ber cheapo 5m USB repeater cables and two 10ft passive cables with a powered 4port and a powered 7 port hub. I need another 20 to 30 feet to add my second telescope this spring. My setup will look like this I am thinking:

PC in study.
--> 5m repeater and 10ft passive to 4 port hub
-->. USB 1.1 cat5 100ft running out to Scope 2 ( future plan)
-->. 5m repeater
-->. 5m repeater to 4 port hub
Scope 1. --> scope control, autoguider and USB 2.0 camera
--> 10 ft passive
-->. 5m repeater
--> 10 ft passive
Scope 2.
USB 2.0 compliant camera
USB 1.1 to scope control


Right now I have this setup to scope 1 wi no issues,

Al


P.S. my super cheapo USB repeaters are:
http://www.amazon.co.../B004I1E6MU/...

#40 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 13 January 2013 - 09:05 PM

Hi Al,

Interesting setup... Hey, if the el cheapo repeater cables work for you, your in business... I was surprised to see the price as low as it is... That is a significant factor...

If it ain't broke, don't fix it!!!

A fellow on another forum posted a Amazon link to
"Alfa USB 2.0 Active Extension Cable - Repeater Cable - High Speed 480Mbps - A Male to A Female - 15 Meters / 49.2 feet Long"

He is apparently using these cables and likes them... I have bookmarked this cable for future use if I decide I need more than what I have on hand.

Alfa USB Active USb2.0 Extension cable

#41 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 14 January 2013 - 12:49 PM

In my previous post I wrote:

I meant to comment in my previous post, I have a couple sets of the USB to Cat-5 adapter dongles on 100 ft Cat 5 cables and they do work however at slower speed than the Active USB extension cables.

I have not attempted to do a benchmark speed test on them as I've had them for a while and since switching to the active USB extension cables, I've simply not had a need to use them...

That would be a simple test to run, You've piqued my curiosity, think I'll give that a try just to see where they fit in the scheme of link efficiency...


Today I have attempted to run a benchmark test of a 100 Ft. Cat-5e cable using the Cat-5e/USB dongle adapters (Also a Tether Tool product) configured as a direct connect from a USB port on the laptop to the Passport HD USB drive and the Passport HD refused to start up. As the Passport HD USB drive gets its power from the USB port, I put a 4 port independently powered hub on the far end of the cable and plugged the Passport HD drive into the hub.

Now the Passport drive powers up, the indicator LED is blinking indicating communications and the Laptop recognizes that the HD drive is connected however does not recognize it as a HD, shows no capacity size, and tells me the drive must be formatted before it can be used.

I then replaced the 100 ft. Cat-5e cable with a 10 ft. Cat-5e cable and repeated the test with the same results.

There was no way that I could get the Laptop to communicate correctly with the Passport HD drive using the Cat-5e dongles and as a result I could not get the HDspeed utility to benchmark the Passport HD USB drive, The HDspeed test utility did recognize the drive was connected however reported the drive incorrectly when using the 10 ft. Cat-5e cable and any attempt to tun a benchmark test came back with 100% errors...

I know the Cat-5e dongles with this 100 ft. Cat-5 cable does work as I have successfully used it with the Canon 60Da camera and to confirm that it was working as expected, I reconnected the USB/Cat-5 dongles to the 100 ft. Cat-5 cable and substituted the Canon 60Da camera for the Passport HD and was able to visually compare the 60Da in both imaging and live view modes same as the tests yesterday of visually comparing the WiFi wireless-N link between the computers.

The test today with the 60Da as a visual comparison would seem to be similar or slightly faster than the results seen yesterday with the wireless-N link test.

Therefore in my scale of results of tests run to date, I would place the USB/Cat-5 dongles over 100 ft of Cat-5e cable as somewhat better than WiFi but not as good as USB over TCPIP using a USB/IP Ethernet hub.

Bottom line is the USB/Cat-5 dongles with 100 ft. Cat-5e cable are definitely usable as a dedicated extension of a USB port on a computer however probably not with a USB hub and most likely results will depend on the type of USB device at the remote end.

#42 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 16 January 2013 - 06:02 PM

Thanks Mitch. It's nice to see you're 'testing' and providing comparisons on the active repeater versus Cat5 USB. Have you thought about writing it up and posting it as an article? I think the community could benefit to have this in 'one place'.

I too came to the same conclusion last year; it's hard to beat active repeater cables.

#43 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 17 January 2013 - 01:20 AM

Thanks Mitch. It's nice to see you're 'testing' and providing comparisons on the active repeater versus Cat5 USB. Have you thought about writing it up and posting it as an article? I think the community could benefit to have this in 'one place'.

I too came to the same conclusion last year; it's hard to beat active repeater cables.


Thank You for your kind comments...

I do agree the information would make a meaningful article however what I have accomplished was done primarily for my own needs and done with a crude make do setup using a freeware "HDspeed" benchmark utility combined with a portable Western Digital USB3.0 HD which was the best combination of what I had on hand to accomplish some sort of evaluation of the different protocols I am interested in using...

I think if this were to be done as an article, proper USB evaluation software should be used to give meaningful readings. The speed results of my testing is meaningless other than being consistent by using the same USB HD device and software, and making the assumption that the software and USB HD device in read mode only, was not affected by running at reduced speeds created by being "choked" over slower communications links other than on what it is designed to be used. With the Western Digital Passport drive plugged into a USB3.0 port on my main computer, it benchmarks consistently at 82M bytes/sec. with 1Mb packet size and in burst mode has hit peaks of 126M Bytes/sec.

That I could not get the Passport drive to work in some communications equipment & link combinations that I tried, I suspect is a limitation of the Passport HD internal operating system. That it would work on the end of 31 meters of active USB cables and not work on a 10 ft length of Cat-5 cables with USB adapter dongles either with or without a externally powered USb hub gives me some doubt.

And yet the same USB adapter dongles on either the 10 ft Cat-5 cable or the 100 ft Cat-5 cable to the 4 port externally powered hub, I was able to operate BYEOS to function with the Canon 60Da camera in either Live View mode or run imaging sessions downloading 20Meg RAW files to the computer. This says to me the USB Cat-5 adapter dongles work reasonably well inspite of the Passport HD refusing to operate over that combination... I see that result as an unanswered question as I have no meaningful speed data to use to determine where the USB/Cat-5 adapters fit in the efficiency scale other than to make a guess based on what I observed that they are slightly better than a WiFi link. But that is a visual comparison by my eye only...

For my own use, I am satisfied with the results and depending on future needs of when and where I setup, I have a good idea of what I will use of the communications gear I have on hand...

I'm happy to share what I have learned, however not sure I have a need to go any further with this testing other than if I can find a better wireless system that improves on the WiFi performance I have experienced so far... I have a need for a Wireless link but none of the WiFi gear I have evaluated so far will support the control and data flow adequately for my needs...

#44 bmwbiker

bmwbiker

    Explorer 1

  • -----
  • Posts: 75
  • Joined: 11 May 2011
  • Loc: SF Bay Area

Posted 27 August 2013 - 03:03 AM

Mitch,
I actually stumbled on this thread while searching on the part number and have essentially the same intent. I don't have a permanent location and am looking to run my whole imaging setup from a portable power tank (No AC power).

Would it be fair to say that as far as you have tested the USBG-4NET hub is meeting up to its specifications?

#45 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 27 August 2013 - 09:41 AM


Mitch,
I actually stumbled on this thread will search on the part number and have essentially the same intent. I don't have a permanent location and am looking to run my whole imaging setup from a portable power tank (No AC power).

Would it be fair to say that as far as you have tested the USBG-4NET hub is meeting up to its specifications?


The USBG-4NET hub is working very well for me altho at this point I have not had a chance to operate with it fully loaded. I am presently using it with my imaging system on the Losmandy G-11 mount which has its own Ethernet interface and in this setup I use the USBG-4NET hub for the imaging camera and the Moonlite focuser USB control. Its working well in this configuration, shows no issues with downloading 22Mb raw images from the imaging camera.

However one interesting development that has turned up, I have acquired a ZWO ASI120MM camera (video, high speed RSB2.0 output) and the ZWO camera will not work with this hub. It possibly is a driver problem with the ZWO camera as other cameras I have work fine including a Orion SSAG camera which is known to be a bandwidth hog in my testing.

And yet I can run a long active USB active extension cable to the ZWO camera and it works fine over the USB extension cable.

This is not a big surprise really, I've known for a long time that autoguiding cameras work best in a dedicated link as they run continuously and use the greatest amount of bandwidth by a factor of 10 in my benchmark testing... But not having the ZWO camera work at all even when testing with a short Ethernet cable (6 ft) on the bench with the hub was a real surprise and I contacted ZWO who's reply was that they had no facility to test the camera over a USB/IP link.

The ZWO ASI120MM camera is the only USB2.0 peripheral device I have found that does not work with the USBG-4NET hub...

#46 ccs_hello

ccs_hello

    Fly Me to the Moon

  • *****
  • Posts: 6548
  • Joined: 03 Jul 2004

Posted 27 August 2013 - 09:46 PM

... The ZWO ASI120MM camera is the only USB2.0 peripheral device I have found that does not work with the USBG-4NET hub...

Perhaps it's using the EZ-USB FX2 chip which is a two-step boot process (load firmware in the first stage.) This definitely will not work in IP-emulating simple USB protocols.

Clear Skies!

ccs_hello

#47 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 27 August 2013 - 10:11 PM

I was never able to get a USB over Ethernet device to work. The only one I didn't try was the Icron Ranger; it's reported to work great from many.

#48 tomcody

tomcody

    Surveyor 1

  • *****
  • Posts: 1778
  • Joined: 06 Jul 2008
  • Loc: Titusville, Florida

Posted 28 August 2013 - 09:42 AM

I was never able to get a USB over Ethernet device to work. The only one I didn't try was the Icron Ranger; it's reported to work great from many.

The Icron Ranger 2204 worked with anything I threw at it including three cameras on continous mode and downloading simutainiously.
The added advantage of the Ranger is industrial strength protectiovn from induced voltage from nearby lightning strikes ( a major cause of failure to remote observatories).
P.S. for at the dome monitoring, I just remoted back from the house computer with either an IPad or laptop. If I wanted to work from the dome, I just unplugged the house computer from the ranger at the dome and plugged in my laptop, everything ran fine.
Rex

#49 fmhill

fmhill

    Messenger

  • -----
  • Posts: 436
  • Joined: 17 Jul 2012
  • Loc: Cape Cod, Massachusetts, USA

Posted 28 August 2013 - 11:41 AM

I was never able to get a USB over Ethernet device to work. The only one I didn't try was the Icron Ranger; it's reported to work great from many.

The Icron Ranger 2204 worked with anything I threw at it including three cameras on continous mode and downloading simutainiously.
The added advantage of the Ranger is industrial strength protectiovn from induced voltage from nearby lightning strikes ( a major cause of failure to remote observatories).
P.S. for at the dome monitoring, I just remoted back from the house computer with either an IPad or laptop. If I wanted to work from the dome, I just unplugged the house computer from the ranger at the dome and plugged in my laptop, everything ran fine.
Rex


When I first started remote controlled imaging, I used a laptop at the mount and as I live in am area (Cape Cod) where night time humidity runs between 93% to 100% typically, I ended up with a $348 dollar repair on a HP Pavilion laptop after about 4 months of operation this way.

At the time I started using 20 meter active USB extension cables however good cables suitable for outdoor use are expensive and work surpringly well and if one uses Tether Tools "Tether Pro" brand, Tether Tools is an American USA supplier to the professional photography market and "Tether Pro" active USB2.0 extension cables are lifetime guaranteed by Tether tools...

However as technology evolves, the latest version of the Losmandy G11 mount Gemini-2 controller has a Ethernet port for mount control and this works extremely well, much better than USB with a serial adapter as most mounts require. Plus considering the cost of a 100 ft Cat-5e cable being 1/4 the cost of a Tether Pro USB2.0 20 meter active extension cable, I was inspired to attempt using USB over Ethernet for the imaging camera USB focuser control and autoguider.

I did look at the Icron Ranger however I found the USBgear-4NET four USB2.0 port hub with a 10-100-1000 Gbit Etherrnet link port and comes with both AC power supply and a DC 7 to 24 volt DC power connector. At the time, the Icron Ranger was selling for over $400, I was never able to get information about DC power operation, and at a cost of $95.00, the USBG-4NET became top of my list of choices to evaluate.

So far, the USBG-4NET unit does all that I expect of it, it handles the mount control USB to serial adapter when I'm using my either of my two Celestron mounts, the 60Da imaging camera in Live View or dowmloadng 22Mb raw files, the USB motoruzed focuser control, and a TEMPerHUM USB temperature and humidity monitor, and all works together just fine. I have also plugged a WD 1Tb Passport HardDisk USB3.0/USB2.0 into it when using the G11 mount and the Serial adapter is not needed. I have also used the USBG-4NET with a Orion SSAG autoguiding camera without trouble...

The fact that the ZWO ASI120MM camera refuses to work with the USBG-4NET hub I am convinced is a peculiarity of the ASI120MM/MC driver... I've also had troubles getting the ASI120MM camera working with at least one software and am reasonably convinced the issue is with the ZWO supplied camera driver. In part, I feel this is because the ASI120MM/MC cameras are a new product appearing on the market mid spring of 2013.

I have communicated with ZWO about communication problems with the ASI cameras and ZWO admits to limited testing facilities for communications... I'm not sure what this means other than ZWO has no explanation...

As to the USBG-4NET hub linked by Ethernet, I am very pleased with its quality and operation for a $95.00 device and if the humidity causes it to fail, I'm much happier to pay $95.00 to replace it than I am to pay $348...

#50 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 5436
  • Joined: 20 Jul 2007
  • Loc: California

Posted 28 August 2013 - 11:41 AM

I just use a laptop at the imaging rig, a wireless Ethernet router and another laptop inside the motorhome. Distance is about 85 feet. I run everything from inside with Remote Desktop. This works well, but I wait until morning to transfer the subs to the inside.
Blueman






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics