ehd question

I think he made an assumption, based on the ERD option, and believed it to be true. But when we sort of cornered him into showing proof, he probably discovered he was mistaken.
 
What you describe is not 10MB up and 10MB down. At any point in time data is traveling in only one direction up to a max of whatever the link will support. You will never get 15MB one direction while also getting 5MB the other direction. USB, or any other half duplex media for that matter, never says more bandwidth this way and less that way. At any point in time it is all bandwidth in one direction Over the length of the transfers it may appear to average out this way but that is not what is going on. Now take into account that we are dealing with video here. Pausing for any real length of time is not an option either during reads or writes. There is only so much buffer to empty before you miss frames on the write or the video stutters or stops on the reads. Video is also an isochronous data flow so certain reservations are made.

The easiest way to show that this is not a viable option on USB 2.0 would be to do the following. Connect and external USB drive to your computer. Mac, Windows, Linux doesn't matter. Now copy two video files to this drive. Next open those files and let them begin playing. Next drag first one file to the hard drive while the two videos are playing, while that transfer is still going drag another over there, then another until you have 4 transfers going while the two videos are playing. Don't drag all four files at one time or it will stack them and not be doing 4 writes while trying to do 2 reads. Are you videos still playing smoothly if at all?

Now if it were full duplex, like USB 3.0 it should work, since the videos playing from the drive would never have to stop streaming to allow the new recordings time to be written to the drive. In a full duplex environment, then USB 2.0 would likely provide enough bandwidth for this type of operation.

Sorry, but what you're trying to say makes absolutely no sense. 802.11g, for example (well even 802.11n, when not using Frequency-division duplexing,) are also half-duplex. That doesn't mean that you can't upload while downloading at the same time.

Just read this:

Duplex (telecommunications) - Wikipedia, the free encyclopedia

Time-Division Duplex (TDD) is the application of time-division multiplexing to separate outward and return signals. It emulates full duplex communication over a half duplex communication link.
Time division duplex has a strong advantage in the case where there is asymmetry of the uplink and downlink data rates. As the amount of uplink data increases, more communication capacity can be dynamically allocated, and as the traffic load becomes lighter, capacity can be taken away. The same applies in the downlink direction.

And this:

Time-division multiplexing - Wikipedia, the free encyclopedia

Time-division multiplexing (TDM) is a type of digital (or rarely analog) multiplexing in which two or more bit streams or signals are transferred apparently simultaneously as sub-channels in one communication channel, but are physically taking turns on the channel. The time domain is divided into several recurrent timeslots of fixed length, one for each sub-channel. A sample byte or data block of sub-channel 1 is transmitted during timeslot 1, sub-channel 2 during timeslot 2, etc. One TDM frame consists of one timeslot per sub-channel plus a synchronization channel and sometimes error correction channel before the synchronization. After the last sub-channel, error correction, and synchronization, the cycle starts all over again with a new frame, starting with the second sample, byte or data block from sub-channel 1, etc.
So again, you are mistaken in saying that it is not possible to do this on USB 2.0. As I said earlier, it's a matter of milliseconds in which the uplink and downlink are switched, therefore, it would have absolutely no affect on video transfer Also, in your EHD example, what you are describing is bandwidth saturation, which occurs when a file transfer attempts to use all available bandwidth while another file transfer is occurring. It is true that video playback may stutter in this scenario, due to increased latency from bandwidth saturation, but it would not be the case when recording only 6.8MB/sec to the EHD versus saturating the USB 2.0 bandwidth.
 
Last edited:
Did you try what I had asked? If it would make you feel any better rather than transferring 4 files do 4 video captures to the HD while playing back 2 Videos. I asked you to try this as it is the easiest way to convey what I am trying to explain to you or I thought it would be. If you would care to get into a very detailed in depth explanation of this I will do so and not need to refer to Wikipedia to do so.

It is you who are mistaken and if you carefully read the wikipedia quotes you are using you will see this.

Time-division multiplexing (TDM) is a type of digital (or rarely analog) multiplexing in which two or more bit streams or signals are transferred apparently simultaneously

I'm not saying that you can't queue Multiple file transfers, I am saying that they can only be traveling in one direction in any point in time. milliseconds or minutes one direction stops while the other direction moves. Stoping and starting too many times and not being able to feed the buffer either of the HD or the receiver itself means something get's skipped somewhere.

Ok how about this. If USB2.0 EHDs could support what you are saying do you not think that Dish would not make a 722 with no drive and charge you the feature fee to use the box as a DVR? How many boxes do you think that they have had to replace since the 722 was released because of failed HDs? Each time one goes belly up, that is shipping each direction, labor for someone to open the box and remove the drive and the cost of the new drive. With all those expenses do you not think that they would make such a box if the feature could be applied without risking performance complaints? Do you not think that there is a technical reason for only offering this feature on Single Tuner boxes?
 
Did you try what I had asked? If it would make you feel any better rather than transferring 4 files do 4 video captures to the HD while playing back 2 Videos. I asked you to try this as it is the easiest way to convey what I am trying to explain to you or I thought it would be. If you would care to get into a very detailed in depth explanation of this I will do so and not need to refer to Wikipedia to do so.

It is you who are mistaken and if you carefully read the wikipedia quotes you are using you will see this.



I'm not saying that you can't queue Multiple file transfers, I am saying that they can only be traveling in one direction in any point in time. milliseconds or minutes one direction stops while the other direction moves. Stoping and starting too many times and not being able to feed the buffer either of the HD or the receiver itself means something get's skipped somewhere.

Ok how about this. If USB2.0 EHDs could support what you are saying do you not think that Dish would not make a 722 with no drive and charge you the feature fee to use the box as a DVR? How many boxes do you think that they have had to replace since the 722 was released because of failed HDs? Each time one goes belly up, that is shipping each direction, labor for someone to open the box and remove the drive and the cost of the new drive. With all those expenses do you not think that they would make such a box if the feature could be applied without risking performance complaints? Do you not think that there is a technical reason for only offering this feature on Single Tuner boxes?

Sorry but your quote about APPARENTLY SIMULTANEOUSLY disproves nothing as it still EMULATES FULL DUPLEX COMMUNICATION. I couldn't have made it any clearer that USB 2.0 is perfectly capable of 6 HD streams at a time.

I already explained (now going on three times) that the most likely limitation here is in fact the CPU of the DVR as it has to encrypt/decrypt the video streams in real time (you can thank the MPAA/RIAA.) USB 3.0 wouldn't make one iota of difference in how fast the CPU is capable of doing this.

So there's your TECHNICAL REASON. There is a BOTTLENECK, but it's NOT USB 2.0.

There is absolutely no difference in transferring 1 file at 20MB/sec or transferring 4 files at 5MB/sec, or 2 files at 5MB/sec up and 2 files at 5MB/sec down. IT'S STILL THE SAME AMOUNT OF BANDWIDTH.

Do you by any chance work for Geek Squad? You seem to be one of those types.

http://www.urbandictionary.com/define.php?term=Geek%20Squad (definitions 3 4 & 5.)
 
Last edited:
I vote with 3HaloODST on the overall bandwidth issue, as well as the likely explanation for the bottleneck. And that assumes Dish firmware is fully interrupt driven on all data channels. Some of these modest limitations (1) make me suspicious that they're scanning ports!

Note 1: I can't even do TWO streams on the EHD connected to either a 612 or 722. It transfers at most one program at a time, and when that is going on, I can't watch anything on the EHD. So current Dish firmware allows a max of only ONE stream, which I think we can all agree is way WAY less than USB 2 bandwidth.
 
Last edited:
I see the ext option on my 722K with L686 Software. I set a timer using it for 1700 CT tonight. We'll see what it does.
 
I see the ext option on my 722K with L686 Software. I set a timer using it for 1700 CT tonight. We'll see what it does.

I will predict the future. My crystal ball tells me... NOTHING!!! NOTHING will happen (at least on the EHD!)

That'll be $19.95, please.
 
Well it is with egg on my face and humble pie on my plate that I deliver my report. It has taken a few days of testing and another day or so dreading taking my ridicule from you guys.

You (well, many of you) were right. I’ll just have to man up and admit it. I was duped by the Dish 'external timer thingy'. I had so many episodes of the same programs on the EHD that I was wrong about new ones being added as the timers expired. The timers showed as being recorded, but in fact they were vapor.

I wanted badly to believe it was working, but only after completely emptying a drive and watching for new events to be recorded, did I find my mistake.

I’m sorry to have been such a pain, but I’m glad I was entertainment for those of you with more experience. Hopefully I can show my face around these parts again, especially since I have just one more topic which I’m sure will be just as controversial - - how I cured a 622's 'overheating syndrome' and constant reboots without using fans or any other extra venting.

Stay tuned…
 
Well it is with egg on my face and humble pie on my plate that I deliver my report. It has taken a few days of testing and another day or so dreading taking my ridicule from you guys.

You (well, many of you) were right. I’ll just have to man up and admit it. I was duped by the Dish 'external timer thingy'. I had so many episodes of the same programs on the EHD that I was wrong about new ones being added as the timers expired. The timers showed as being recorded, but in fact they were vapor.

I wanted badly to believe it was working, but only after completely emptying a drive and watching for new events to be recorded, did I find my mistake.

I’m sorry to have been such a pain, but I’m glad I was entertainment for those of you with more experience. Hopefully I can show my face around these parts again, especially since I have just one more topic which I’m sure will be just as controversial - - how I cured a 622's 'overheating syndrome' and constant reboots without using fans or any other extra venting.

Stay tuned…

Don't feel too bad. In fact, you gave some of us a challenge, something to "mess with". Since the pocketdish is such a thing of the past, it would be nice if they did away with that confusing option. But since there are still some out there, it will probably remain. So, wipe the egg off your face and tell us about your over-heating solution (you just might want to make dead sure it works this time).:)
 
Sorry but your quote about APPARENTLY SIMULTANEOUSLY disproves nothing as it still EMULATES FULL DUPLEX COMMUNICATION. I couldn't have made it any clearer that USB 2.0 is perfectly capable of 6 HD streams at a time.

I was trying to point out that it was giving the appearance of being Simultaneous. Emulating full duplex is not the same as being full duplex. Emulate is to strive to be like something not to be the same. As an example if you run an application in emulation does it perform as well as it does in a native environment?

The only thing that you have made clear in your arguments is that you are capable of quoting snippets of Wikipedia and often snippets that are dis-related to the topic of USB 2.0 While USB 2.0 does employ Time Division Duplexing (TDD) it does not employ Time Division Multiplexing (TDM)

I already explained (now going on three times) that the most likely limitation here is in fact the CPU of the DVR as it has to encrypt/decrypt the video streams in real time (you can thank the MPAA/RIAA.) USB 3.0 wouldn't make one iota of difference in how fast the CPU is capable of doing this.

No, you have not explained this, you have simply stated it. Explain: Make (an idea, situation, or problem) clear to someone by describing it in more detail or revealing relevant facts or ideas.

The MPAA/RIAA has nothing to do with the conversation at hand, which is the ability of USB 2.0 to handle this task. As to your argument that the limitation is with the CPU, here is another test you can perform. Using your 622/722 record 4 streams while watching 2 from the internal drive. This can be done and therefore dispels your argument that the cpu is the bottleneck.

So there's your TECHNICAL REASON. There is a BOTTLENECK, but it's NOT USB 2.0.

There are many technical reasons USB 2.0 is not up to this task and I am attaching the USB 2.0 doc. If you are so inclined look it over, feel free to ask me any questions you may have via PM and I will try to answer any questions you may have. Pay especial attention to chapter 5 I am not aware of your, or others here, technical backgrounds and as such have tried to keep my replies at a level that any might understand. Providing simple test any could try to verify my assertions. I will not pretend to know all about everything. One of the reasons I am here, to learn about satellite. A topic that I am far from versed on.

There is absolutely no difference in transferring 1 file at 20MB/sec or transferring 4 files at 5MB/sec, or 2 files at 5MB/sec up and 2 files at 5MB/sec down. IT'S STILL THE SAME AMOUNT OF BANDWIDTH.

No, there is a difference. This can easily be seen by means of a simple test using your personal computer and an external HD connected via USB 2.0 First Transfer one file to the drive of say 2GB. Time it. Now drag two files each of 1GB and time this. Times should be very close as the OS when you drag two at once will queue them to be consecutive transfers. Now drag one and while it is transferring drag the other over. This is a bit tougher to get an over all time of because of the lag between the start of the first and you starting the second but it can be done. You will now see a noticeable difference. Repeat this process with a single 4GB file and then 4 1GB files. You will see that the more concurrent transfers you have running the overall time increases.

I will give you the point that no matter what the pipe (physical media) will always have the same theoretical bandwidth, but real world usable bandwidth depends on many factors. Number of tasks. Type of task, control, interrupt, isochronous, bulk. Just a couple of the considerations to be weighed to determine real usable bandwidth. Bottom line USB 2.0 is half duplex and no matter how many ways you slice and dice it data can only flow in one direction at any given moment in time.

Here is something else you can try to demonstrate how the more ways you slice up the over all bandwidth how performance decreases. Take your garden hose, fill a gallon jug and time it. Think of this as data flowing from the receiver to the EHD. Now put a jug behind you and alternate filling both, alternating between them turning the hose off and on as you do. Did it take twice as long or longer. Now do this with 2 jugs behind you, and 4 in front of you. Again not filling one then the next, but rater rotating between all of them turning the hose off as you move from one to the next and then back on. Did this take 6 times as long as just putting the hose in the one jug and letting it run full stream till complete? This is a very highly simplified example but should help to illustrate what is going on.


Do you by any chance work for Geek Squad? You seem to be one of those types.

Sir, and I apologize if I've miss-guessed your gender, personal attacks do nothing to further your position. You have no idea as to my technical background, as I yours. This is the reason I did not go into great technical discussion of this topic. If after looking over my post here and the attached document you have questions or wish to discuss this further I will be very happy to do continue this conversation via PM as I do believe that we have greatly deviated from the original post.
 

Attachments

  • usb_20.pdf
    5.7 MB · Views: 123
Well it is with egg on my face and humble pie on my plate that I deliver my report. It has taken a few days of testing and another day or so dreading taking my ridicule from you guys.


I thought I had discovered the same thing last year, only to be told it was for the Pocket Dish. So, until they removed that from the menu, you aren't the first nor will you be the last to think they had discovered something new.
 
While USB 2.0 does employ Time Division Duplexing (TDD) it does not employ Time Division Multiplexing (TDM)

Sorry, but this alone shows that you really don't know what you're talking about. Time Division Duplexing IS Time Division Multiplexing.



The MPAA/RIAA has nothing to do with the conversation at hand, which is the ability of USB 2.0 to handle this task. As to your argument that the limitation is with the CPU, here is another test you can perform. Using your 622/722 record 4 streams while watching 2 from the internal drive. This can be done and therefore dispels your argument that the cpu is the bottleneck.

Again, proving your ignorance, because the 622/722 do NOT encrypt recordings internally. They DO encrypt recordings to the EHD. Also, 622/722 are NOT capable of recording 4 at once. That is the 722K that does that.


I am not aware of your, or others here, technical backgrounds and as such have tried to keep my replies at a level that any might understand.

You seem to be the only one here with problems understanding anything.


I will give you the point that no matter what the pipe (physical media) will always have the same theoretical bandwidth, but real world usable bandwidth depends on many factors.
Well no crap. That's what I have been saying all along. 20MB/sec is more than enough for 6 HD streams. Even if it is split up and shared between recording/playback.

Bottom line USB 2.0 is half duplex and no matter how many ways you slice and dice it data can only flow in one direction at any given moment in time.

Great job arguing semantics, but that still doesn't disprove the fact that USB 2.0 is capable of handling the task at hand.

Take your garden hose, fill a gallon jug and time it.
Think of this as data flowing from the receiver to the EHD. Now put a jug behind you and alternate filling both, alternating between them turning the hose off and on as you do. Did it take twice as long or longer. Now do this with 2 jugs behind you, and 4 in front of you. Again not filling one then the next, but rater rotating between all of them turning the hose off as you move from one to the next and then back on. Did this take 6 times as long as just putting the hose in the one jug and letting it run full stream till complete? This is a very highly simplified example but should help to illustrate what is going on.
Now you're just being ridiculous. A USB connection is not a garden hose. Water can't change directions in a matter of milliseconds. Your point is invalid.

Sir, and I apologize if I've miss-guessed your gender, personal attacks do nothing to further your position. You have no idea as to my technical background, as I yours. This is the reason I did not go into great technical discussion of this topic. If after looking over my post here and the attached document you have questions or wish to discuss this further I will be very happy to do continue this conversation via PM as I do believe that we have greatly deviated from the original post.

Attaching a document doesn't prove anything about your technical knowledge. Why don't you show me where in that 650-page document, it says that since it is half-duplex, it is incapable of moving data in two directions simultaneously (or NEAR SIMULTANEOUSLY?)
 
Sorry, but this alone shows that you really don't know what you're talking about. Time Division Duplexing IS Time Division Multiplexing.

This again is incorrect. Duplexing and Multiplexing are different beasts. Duplexing relates to sending data in different directions. Multiplexing involves sending data from multiples sources or to multiple destinations within a given transaction.

I believe what you may have intended is that USB employs STDM (statistical time division multiplexing), which is not the same as TDM, which is not the same as TDD.

Again, proving your ignorance, because the 622/722 do NOT encrypt recordings internally. They DO encrypt recordings to the EHD. Also, 622/722 are NOT capable of recording 4 at once. That is the 722K that does that.

To this I admit I am in error and that the 722k is the only unit with this ability. I did however say I do not claim to be an expert in satellite which is the reason for my being here, to learn. I do have a 722k however did not notice the "K" designation. Of this you are right, and I humbly stand corrected.

Well no crap. That's what I have been saying all along. 20MB/sec is more than enough for 6 HD streams. Even if it is split up and shared between recording/playback

You missed my point here. While the theoretical bandwidth of the pipe is there, there is more to bandwidth than just the physical media and the realworld, usable bandwidth will vary depending on what we are attempting to do with that physical media.

Now you're just being ridiculous. A USB connection is not a garden hose. Water can't change directions in a matter of milliseconds. Your point is invalid.

And did I ever say that an USB connection was a garden hose? I simply gave this example to show just some of what is going on in doing multiple USB transactions and multiple directions. Now if you really want to get technical about this I could point out that USB could change direction in 1 millisecond which is the frame length.

Attaching a document doesn't prove anything about your technical knowledge. Why don't you show me where in that 650-page document, it says that since it is half-duplex, it is incapable of moving data in two directions simultaneously (or NEAR SIMULTANEOUSLY?)

Nowhere in that document will you find that USB can not move data in two directions near simultaneously, but that is not the question at hand. Now a few of things that you will find in that document are descriptions of the different types of transfers (Control, Interrupt, Isochronous, and Bulk). That video would be an isochronous transfer. That no more that 3 isochronous transfers may be contained in a micro frame (125 nanoseconds or 1/8 millisecond) allowing a data rate up to 192Mb/s or 24MB/s (this does not include overhead).

The examples that I gave for all to try who were interested in seeing that as you add enpoints (tasks if you will) the usable bandwidth decreases, and this is best case as these are Bulk transfers. While according to your assertion if I can transfer 1 4GB file in say 1 minute I should then be able to transfer 4 1GB files concurrently in 1 minute as well.

This will be my last public post to this topic, as again, we have deviated greatly from the original post. As I said previously though I am happy to continue this conversation or try to answer additional questions you or other may have on this subject in PM.
 
This again is incorrect. Duplexing and Multiplexing are different beasts. Duplexing relates to sending data in different directions. Multiplexing involves sending data from multiples sources or to multiple destinations within a given transaction.

TDD IS TDM. "Multiple sources" = uplink + downlink.

I believe what you may have intended is that USB employs STDM (statistical time division multiplexing), which is not the same as TDM, which is not the same as TDD.

STDM is a TYPE of TDM. It's in the name, not that hard to figure that out. TDM is what is used to employ TDD. So TDM and TDD ARE THE SAME. Common sense.

To this I admit I am in error and that the 722k is the only unit with this ability. I did however say I do not claim to be an expert in satellite which is the reason for my being here, to learn. I do have a 722k however did not notice the "K" designation. Of this you are right, and I humbly stand corrected.

It's certainly not your first error. You completely skipped over the fact that it does NOT encrypt recordings recorded to the internal HDD. It DOES encrypt recordings placed on the EHD. You said in a previous point that the MPAA/RIAA have nothing to do with it when in fact they have everything to do with this as if it weren't for them DISH wouldn't have implemented encryption on the EHDs.



You missed my point here. While the theoretical bandwidth of the pipe is there, there is more to bandwidth than just the physical media and the realworld, usable bandwidth will vary depending on what we are attempting to do with that physical media.
You can put it any which way you like, but the bottom line is 6 HD streams aren't that demanding (given the current bit rates) and USB 2.0 is NOT the limitation. It IS the encryption. Why is that so hard for you to figure out?



And did I ever say that an USB connection was a garden hose? I simply gave this example to show just some of what is going on in doing multiple USB transactions and multiple directions. Now if you really want to get technical about this I could point out that USB could change direction in 1 millisecond which is the frame length.

Obviously you haven't read anything I've posted, as I've been saying all along that the data flow changes directions in milliseconds. You act as if quoting Wikipedia is meaningless, when in fact it has sources to back up it's claims, and clearly you never read any of the quotes I posted.



This will be my last public post to this topic, as again, we have deviated greatly from the original post. As I said previously though I am happy to continue this conversation or try to answer additional questions you or other may have on this subject in PM.

Good. I've been getting tired of saying the same thing over and over, because you obviously don't read. It must hurt your Geek Squad ego to just admit that you are wrong. All of your posts haven't proved anything. You think that just throwing around terms and throwing 650-page documents at me proves that you know more on the subject, yet you do nothing to explain how in fact this proves your point. You then go on to create ridiculous analogies that aren't even remotely helpful. They give me a good laugh though.
 
Well it is with egg on my face and humble pie on my plate that I deliver my report. It has taken a few days of testing and another day or so dreading taking my ridicule from you guys.

You (well, many of you) were right. I’ll just have to man up and admit it. I was duped by the Dish 'external timer thingy'. I had so many episodes of the same programs on the EHD that I was wrong about new ones being added as the timers expired. The timers showed as being recorded, but in fact they were vapor.

I wanted badly to believe it was working, but only after completely emptying a drive and watching for new events to be recorded, did I find my mistake.

I’m sorry to have been such a pain, but I’m glad I was entertainment for those of you with more experience. Hopefully I can show my face around these parts again, especially since I have just one more topic which I’m sure will be just as controversial - - how I cured a 622's 'overheating syndrome' and constant reboots without using fans or any other extra venting.

Stay tuned…

Now you really ARE one of us! We've all made mistakes of one sort or another. Welcome to our even tighter embrace.
 
***

Users Who Are Viewing This Thread (Total: 1, Members: 0, Guests: 1)

Who Read This Thread (Total Members: 1)