|
|
Post by artiek on Apr 20, 2010 22:17:52 GMT -5
But I wanted one! Now I think you're saying it's not going to get any better? Is that possible?
|
|
|
|
Post by sbavnut on Apr 20, 2010 23:16:00 GMT -5
A couple of days ago, someone quoted Vince as saying something like this - "contrary to speculation in the forum, the UMC-1 is capable of acting as a USB Host. Hence a USB stick update is possible."
The next time someone calls Emotiva, can they please verify this? Also, even if the UMC-1 is capable of acting as a USB Host, is Emotiva willing to spend the time and money to deliver such a mode of updating?
|
|
|
|
Post by monkeypimp on Apr 20, 2010 23:41:48 GMT -5
A couple of days ago, someone quoted Vince as saying something like this - "contrary to speculation in the forum, the UMC-1 is capable of acting as a USB Host. Hence a USB stick update is possible." The next time someone calls Emotiva, can they please verify this? Also, even if the UMC-1 is capable of acting as a USB Host, is Emotiva willing to spend the time and money to deliver such a mode of updating? That was me that said that and he did say that. But someone today said they spoke to Lonnie and he said it wasn't. I think there is some confusion at the Emotiva headquarters, which makes it even more confusing for us. |
|
|
|
Post by BillBauman on Apr 21, 2010 0:31:57 GMT -5
A couple of days ago, someone quoted Vince as saying something like this - "contrary to speculation in the forum, the UMC-1 is capable of acting as a USB Host. Hence a USB stick update is possible." The next time someone calls Emotiva, can they please verify this? Also, even if the UMC-1 is capable of acting as a USB Host, is Emotiva willing to spend the time and money to deliver such a mode of updating? That was me that said that and he did say that. But someone today said they spoke to Lonnie and he said it wasn't. I think there is some confusion at the Emotiva headquarters, which makes it even more confusing for us. Hey, guys, I'm the other 'someone' here. I think Vince may have been unclear as to which processor either was asked about or he'd been briefed on. He's probably still trying to wrap his head around all the technology at Emotiva and he certainly knows about a ton of products and future Pre/Pros that we don't. That said, I had a very nice conversation with Lonnie and he was the one who actually brought up the fact that the UMC-1 does not have a USB host controller. We were discussing possible alternative methods of doing the firmware update. Believe me, for now, we have what we have. I honestly don't think it's the end of the world. That doesn't mean the alternatives won't continue to get explored, but it does mean they are limited and may be costly. |
|
avnut
Minor Hero
Posts: 43
|
Post by avnut on Apr 21, 2010 8:00:13 GMT -5
Bill, Thanks for the VERY informative post. I think that Emotiva probably made a good decision on the update method in order to keep the cost and hence the price very attractive. The missing element here is that this approach also requires a very effective Beta test program so the probability of requiring firmware updates by customers is minimized. My solution to this is to wait until the UMC-1 has proven to be stable for a while (my thanks to all of you beta testers) or to wait for the XMC-1. In my case impatience has proven once again not to be a virtue while the fact that I am an upgrader and hence did not get an early UMC will prove to have more value than just the discount. Bill, the time you have taken to help with this upgrade process and to explain it clearly is greatly appreciated by me. Thank you again.
|
|
|
|
Post by sbavnut on Apr 21, 2010 10:09:49 GMT -5
Well - that answers the "UMC-1 as a USB Host" questions.
Bill - thank you very much for taking the lead and getting the correct answers.
I just hope that the current process could be made "better" (somehow - at the very least, add some post update checksum process), given the constraints of time and money.
|
|
|
|
Post by ÈlTwo on Apr 21, 2010 10:10:36 GMT -5
Bill, Thanks for the VERY informative post. I think that Emotiva probably made a good decision on the update method in order to keep the cost and hence the price very attractive. . . . . Bill, the time you have taken to help with this upgrade process and to explain it clearly is greatly appreciated by me. Thank you again. +1 |
|
|
|
Post by BillBauman on Apr 21, 2010 10:37:41 GMT -5
I just hope that the current process could be made "better" (somehow - at the very least, add some post update checksum process), given the constraints of time and money. I'm waiting on some documentation at the moment. If I can figure out a way to ask the microcontrollers 'questions' (query them), I may be able to work out a poor man's checksum. It's a long shot, but if I make any progress, everyone will hear about it. |
|
RSavage
Emo VIPs 
My goal is to live forever. So far, so good.
Posts: 674
|
Post by RSavage on Apr 21, 2010 11:25:32 GMT -5
I'm waiting on some documentation at the moment. If I can figure out a way to ask the microcontrollers 'questions' (query them), I may be able to work out a poor man's checksum. It's a long shot, but if I make any progress, everyone will hear about it. Checksums would make this old boy very happy. R |
|
ICBM99
Emo VIPs
When will then be now? ...Soon.
Posts: 1,702
|
Post by ICBM99 on Apr 21, 2010 11:28:50 GMT -5
I'm waiting on some documentation at the moment. If I can figure out a way to ask the microcontrollers 'questions' (query them), I may be able to work out a poor man's checksum. It's a long shot, but if I make any progress, everyone will hear about it. Checksums would make this old boy very happy. R +1 Except the old part.  ;D And were talking about checksums during the firmware update right, and not just downloading.  |
|
|
|
Post by brettjb on Apr 21, 2010 11:34:09 GMT -5
And were talking about checksums during the firmware update right, and not just downloading. I suspect Bill is talking about checksums post-update Basically, you run the update, then query the processors to calculate their "checksums". Compare that to the known good values. If it matches, you're good. If not, re-flash. --Brett |
|
ICBM99
Emo VIPs
When will then be now? ...Soon.
Posts: 1,702
|
Post by ICBM99 on Apr 21, 2010 11:35:22 GMT -5
That's what I thought, just making sure. ;D
|
|
|
|
Post by BillBauman on Apr 21, 2010 11:56:59 GMT -5
And were talking about checksums during the firmware update right, and not just downloading. I suspect Bill is talking about checksums post-update Basically, you run the update, then query the processors to calculate their "checksums". Compare that to the known good values. If it matches, you're good. If not, re-flash. --Brett Yeah, pretty much. Although, even if it is possible to establish communication, it's seldom as easy as one might like it to be. It's probably more like, ask, make a guess, ask, make a guess, ask, hope for the best, make up a number, stand on your head, then tell the user everything's fine, even though nothing was actually confirmed or denied. ;D |
|
|
|
Post by hollip3020 on Apr 21, 2010 12:49:48 GMT -5
Bill, thank you for clearing this up, and I understand the streams could get crossed (I've run into this issue before with a few things I've developed), but I still don't see how the whole process can't be automated to some degree. Even with the cookie cutter USB stack/driver, it still communicates with at least the front panel controller to some degree, which means it could be made to issue commands such as power on/off, etc. if it doesn't already. What this boils down to is the firmware update application could handle uploading any necessary files and all front panel button presses. The only user input that would then be necessary, would be the initial starting of the process with a single click and the power cycle required after the first file has been loaded. That coupled with disabling the front panel buttons during the process would effectively eliminate all user error problems save physically flipping the power switch on the back. As a software engineer with my fair share of hardware interfacing and UX design/development, it would have never been an option to require the user to upload each file to the unit individually and perform the necessary front panel button presses. There are just too many variables present to ensure an acceptable success rate. Unfortunately, I don't have a UMC-1 yet (still waiting on that fateful email) and can't even pretend I know the full capabilities of the each controller in the unit, but even if the front panel button presses can't be automated, the file uploads could still be, and any other user actions could be plainly specified at the required time by the update application. With all this said, I don't mind undergoing processes such as this as much as my less tech savvy brethren, and I know this thread is mainly about the shortfalls of saving a little money on a USB controller, but I find it incredibly hard to believe there is no way the whole process couldn't be improved. Sure, the method I've described won't make up for the shortcomings of the USB controller, but, and I mean this in the most non-offensive way, it will make up for the shortcomings of its users. I stand behind Emotiva and I respect them for making such awesome gear for such low prices, it's just unfortunate that, in some people's minds, the few mistakes they've made over the last few months have outweighed years of outstanding hardware. P.S. If this method has already been discussed and proved impossible given the hardware restrictions, I'm sorry for wasting your time. I don't have the time or motivation to read through every thread on this matter to make sure my idea hasn't been posed yet. P.P.S Dan and Lonnie, I'd love to help you out with testing the next firmware, I just need a UMC-1 first!  |
|
|
|
Post by BillBauman on Apr 21, 2010 13:33:10 GMT -5
hollip3020, I hope I didn't make anything sound impossible. Everything you've said is certainly possible, but again, for a price. I think you'll have the lights-come-on moment (hopefully) when I explain why this likely wasn't done.
You're a developer, so, this should make sense. You have a process that needs to take place, you plan it to happen maybe 2 or 3 times, 5, tops. The process accesses any of up to 5 different thingies, let's just call them chips. Each time, it may need to access 1 or 2, or all 5 thingies, but you don't know until you're further down the development path. Now, you don't have the in-house resources to create the tool to perform the process, and you're trying to keep costs to an absolute bare minimum. Do you?
A) Contract a company to create a tool that is not really all that automated, but can perform the process under any of the possible circumstances, albeit perhaps with some (significant) user interaction.
B) Contract a company each time you get to the point of needing to perform the process, and pay them each time to completely automate it, whereby making each process a significant cost, as well as adding significant delay to the start time of the process because the exact required functions of the automated tool are not known each time until the process needs are completed.
If you picked A, then you would likely fit in on the Emotiva development team. If you picked B, then you would like to pay more for the UMC-1 or pay for each firmware update. (that one would certainly have people scratching their heads)
[speculation]
There is also one other (many others, really) technical aspect of this I have not addressed. That is, until I dig a little deeper, I won't know if this entire solution is sort of a Windows-only attached system. Much like original soft-modems (for those that remember), AKA - WinModems. These were traditional dial-up modems (most of us still have one in our laptops) that do not have discrete processors but rather rely on a firmware/driver approach to function. When they were first shipped, Linux (and any other OS) users could not use them (until open source and alternative platform drivers were written). Or, like modern motherboard-integrated "RAID" controllers. The Linux community has finally given up on those and refuses to acknowledge them or even attempt to mask their "RAID-like" functions. The Silicon Lab solution may likely be a Windows-only solution (depending on which IC was used) and would require significant dev work to be made to work in a non-Windows world.
[/speculation]
|
|
|
|
Post by agehring on Apr 21, 2010 13:42:46 GMT -5
I can sum this up in one sentence:
Fast, cheap, good; pick two!
__________________________
Good + Fast = Expensive
Good + Cheap = Slow (I think the is (should be?) Emotiva's Goal)
Fast + Cheap = Inferior
You truly get what you pay for!
|
|
avnut
Minor Hero
Posts: 43
|
Post by avnut on Apr 21, 2010 13:44:05 GMT -5
Nothing is impossible, but it all has to do with cost. I am going to do a little poking around to determine the scope of exactly this, as I'm an advocate of this sort of approach. Don't get your hopes up, it isn't as simple as open, write, close. Bill, Your explanation was very good as I said in an earlier post but let me ask another question. This will probably show my ignorance but I will ask it anyway. In my equipment, the gold standard for updating is my PS3. Of course it has an Ethernet interface. The PS3 automatically checks for updates, asks if I want to download and install and if I say yes, it does the rest. The PS3 currently only costs about $300 (I think) and I would expect that the cost of the update facility is only a small part of the total cost. Of course, Sony has to have a dedicated system for processing updates. Are the systems in the UMC so different or so much more complex than software in the PS3 that such a system would not have been feasible? An obvious difference, of course, is that the PS3 was designed to be updated frequently, and I suspect that maybe the UMC was not expected to need updates after two or three initial updates. Anyway, there are probably very good reasons for designing the UMC the way it was but I am confused by why a relatively cheap machine like the PS3 can have such a good update system. Ray |
|
|
|
Post by BillBauman on Apr 21, 2010 14:03:56 GMT -5
Nothing is impossible, but it all has to do with cost. I am going to do a little poking around to determine the scope of exactly this, as I'm an advocate of this sort of approach. Don't get your hopes up, it isn't as simple as open, write, close. Bill, Your explanation was very good as I said in an earlier post but let me ask another question. This will probably show my ignorance but I will ask it anyway. In my equipment, the gold standard for updating is my PS3. Of course it has an Ethernet interface. The PS3 automatically checks for updates, asks if I want to download and install and if I say yes, it does the rest. The PS3 currently only costs about $300 (I think) and I would expect that the cost of the update facility is only a small part of the total cost. Of course, Sony has to have a dedicated system for processing updates. Are the systems in the UMC so different or so much more complex than software in the PS3 that such a system would not have been feasible? An obvious difference, of course, is that the PS3 was designed to be updated frequently, and I suspect that maybe the UMC was not expected to need updates after two or three initial updates. Anyway, there are probably very good reasons for designing the UMC the way it was but I am confused by why a relatively cheap machine like the PS3 can have such a good update system. Ray Ray, you certainly picked an interesting one for comparison. This is not an easy answer. First off, the PS3, to this day, loses money every time Sony sells one. Much like HP printers are loss-leaders designed to make up the profit in selling ink, the PS3 is a loss-leader designed to make up the profit in selling software titles and services for it. The UMC-1 has neither of these go-to-market strategies, it needs to be profitable on its own, unless we all plan to pour Emotiva ink cartridges on it and buy all of our cables and amplifiers for it exclusively from Emotiva (I know, some of you are already clamoring for the Emo Ink preorder list). That said, let's pretend the PS3 wasn't a loss leader, let's just talk technology. The PS3 is a fully functional computer. It has an entire operating system on it that has been amortized in cost over many generations of consoles and consumer electronics. Sony has very big pockets and can afford to lose money it for years before they become profitable. They can also over-design the hell out of it and put thousands of developers to work on it. The UMC-1 doesn't even have a USB master controller hub, much less an independently functioning, stand-alone operating system. These sort of things will likely come in future products from Emotiva, but let's say they tried to deliver all of that in the UMC-1, regardless of cost. I think we'd all agree the UMC-1 is complex enough as it is. Does anyone here seriously want to be working out the kinks of a first-gen integrated operating system along with everything else we're dealing with right now? The PS3 is beyond a 3rd-generation platform with probably billions of dollars in development over its lifetime. The UMC-1 is a gen 1 platform with at least 20 bucks in development dollars. Ok, just kidding, but, seriously, the dev dollars alone for the PS3 could probably buy all of the product Emotiva's ever sold. Comparing where the two companies' products are today in their life cycle is unfair. I can already hear the folks talking about Oppo or Gizmo or Streamo or Whammo. Every company has a product, a cost to it, design decisions to make and live with. This is where we are with the UMC-1 today. We can all find examples of better and worse, when it comes to the update process. This isn't directed at just this particular post, but really, at some point we all just have to accept what is and make the best of it. I know it's not ideal, but by supporting each other, I think we have all made it 10 times easier than it was when we all started. |
|
|
|
Post by BillBauman on Apr 21, 2010 14:15:32 GMT -5
Ray, I should have separated my previous response into two posts. I do not mean to sound like I'm calling you out or calling your question stupid. I perfectly understand that you stated up front you were just asking about it and that you did not have a clear understanding of the differences. I did want to make the point, though, that we are where we are, and we'll (probably) all get through it (eventually).
;D
|
|
avnut
Minor Hero
Posts: 43
|
Post by avnut on Apr 21, 2010 15:01:54 GMT -5
Bill, That, as usual, was helpful. We all tend to look at one thing or another we have and ask ourselves why something else can not work like that. I am looking forward to the UMC. I am glad we only have to pay $699 or whatever it turns out to be. I, like a few others, here just kept asking myself why can't they fix the darn thing, or why does it have to be so complicated. I have a much better appreciation of the product we are working with. After all I do want a Pre/pro and not another fully functioning computer. I will just wait until the kinks have been ironed out and then order one. If the price is more then, I am sure it will still be a good deal. Or more probably, at this point, I will just wait for the XMC. Bill, you are a great communicator. Thanks for taking the time to educate me.
Ray
|
|
;D 
