On Sun, Apr 25, 2010 at 7:16 AM, Drax Felton <[email protected]> wrote: > I disagree with this statement. It's not a waste of development to make the > Flex software deal with this issue smoothly. > > Flex is a commercial product that has to live in the real world of > mainstream, Windows based computers, all their faults included.
Perhaps we have a different point-of-view on this. My take is very simple: when there is a problem, fix the problem; don't apply a band-aid. In this case it is not possible to fix the problem in PowerSDR. Yes, it is possible to make recovery more graceful but how many people are going to be happy with a top-of-the-line radio that periodically goes into audio crap-out no matter how fast it recovers? And as long as you are speaking in terms of business, what functional feature are you willing to give up in order to have more rapid recovery from raspiness? Me? None. PowerSDR has its own set of problems that need work, all of them higher in my priority list than anti-raspiness, especially since *I* can fix that problem by tuning my PC properly or, in the extreme case of crappy-PC, getting a proper PC. And here is the thing: this is not just a Windows problem. I was discussing this with someone who writes drivers for Linux and he pointed out that Linux is going to have the same kinds of problems with the same hardware. You can ameliorate the problem by doing a really good job of writing your drivers and keeping the atomic portions of the interrupt handler as short as possible so that you can dispatch other interrupts quickly, but that doesn't change the fact that the problem exists in the hardware. There is only so much that software can do to accommodate limitations in hardware. I like to drive fast on a race track and I like to do aerobatics. I do these things with hardware that has been designed and prepared for the task. I do not take a random car or airplane and then try to do something with them that they are not designed to do. You have chosen to do SDR with a computer. You need proper hardware to ensure that the job gets done properly. Why don't I hear people complaining to Icom that their IC-718 isn't doing the job when contesting? Hams don't even think twice about spending more money to get a rig that will do the job that they want to do -- contesting, digital modes, EME, weak-signal tropo, etc. No one expects the IC-718 to do the job of an IC-7800 ... or a Flex 5000 for that matter. So why do you expect a $300 Wal-Mart special PC to be wholly adequate to be the other half of your SDR? The Wal-Mart special costs $300 BECAUSE THEY CUT CORNERS! Why do you expect Flex to bend over backwards to try to force hardware that is poorly designed to do a job it is not capable of doing ... when they don't even make that hardware in the first place? Hey, I betcha that the Flex 5000c doesn't have these problems. Flex ensured that 'C' had proper hardware. But you, the consumer, wanted to save money and bought the Flex 5000a or the Flex 3000. It is now up to you, not Flex, to ensure that the supporting hardware YOU supply is up to the task. Don't blame Flex because YOU chose to cut corners. And, yes, I know that there are problems with some "high-end" computers. The reason that their manufacturers didn't bother to make sure there will be no DPC problems is that ... the market hasn't demanded it. When manufacturers find that people won't buy their hardware with these limitations, they will remove the limitations in order to maintain sales. In the mean time you are going to have to be an educated consumer to get a tool that is appropriate to the job at hand. I bet you do that for all the other stuff in your shack anyway. Why balk at this? Different hardware has different performance. By definition better hardware has better performance. The computer is half of your radio's hardware. If you use a crappy computer you are going to have a half-crappy radio. Expect half-crappy performance. Period. Now, do you want to bitch or do you want to fix the problem? Your choice. There are quite a few people here who will be happy to assist you if you chose to solve the problem. -- 73 de Brian, WB6RQN/J79BPL _______________________________________________ Flexedge mailing list [email protected] http://mail.flex-radio.biz/mailman/listinfo/flexedge_flex-radio.biz This is the FlexRadio Systems e-mail Reflector called FlexEdge. It is used for posting topics related to SDR software development and experimentalist who are using alpha and beta versions of the software.
