Are you happy with Sage dependability?

[thatdude90210]

Mine is a dedicated Sage server / HD tuner-recorder (myhd130) that's very stable. It's on 24/7 and rarely needs a reboot ... only when I update something. I don't use my HTPC to play games or browse the internet.

[Lucas]

Mine is very reliable. It runs 24x7 and I only reboot it once a week or so.

I do notice however, that things like stuttering are more frequent with the betas.

I take special care on what software I install on it and always have backup images.

I think that the stability is directly related to the hardware.

[elaw]

Originally Posted by jominor

It is mathematically IMPOSSIBLE to create software without bugs.

I'm with lotusvball on this one, what you say simply ain't true. Every TV set built today has one or more microprocessors in it, when was the last time you had to reboot your TV?

I think part of the issue here is what is defined as a "bug". If I'm given a software spec that says "program shall add two integers with values from 0 to 100 and display the result", I can write a program that will perform according to spec 100.0% of the time. If the spec is more complex/vague, like "record TV programs reliably on a PC with an infinitely variable software and hardware environment", things get a little more difficult. And technically if there's no spec at all, you can't call anything a bug!

However I also agree with some other posters here that a big part of the problem is feature creep. Can't tell you how many times I've been tempted to start a tongue-in-cheek thread on the beta forum titled "Can we get placeshifting on a #2 Philips screwdriver?". I really wish the Sage guys would pay more attention to either making the software more tolerant of hardware/software variations, or identifying and making known which hardware/software can be "trusted" and which can't.

One other thing I've noticed that might be interesting to pursue: it seems to me that most (not *all* but most) of the posters on the forums who are happy with Sage's stability are not recording HD, and vice versa. Anyone else noticed this?

Eric

[jominor]

Originally Posted by elaw

I'm with lotusvball on this one, what you say simply ain't true. Every TV set built today has one or more microprocessors in it, when was the last time you had to reboot your TV?

I think part of the issue here is what is defined as a "bug". If I'm given a software spec that says "program shall add two integers with values from 0 to 100 and display the result", I can write a program that will perform according to spec 100.0% of the time. If the spec is more complex/vague, like "record TV programs reliably on a PC with an infinitely variable software and hardware environment", things get a little more difficult. And technically if there's no spec at all, you can't call anything a bug!

However I also agree with some other posters here that a big part of the problem is feature creep. Can't tell you how many times I've been tempted to start a tongue-in-cheek thread on the beta forum titled "Can we get placeshifting on a #2 Philips screwdriver?". I really wish the Sage guys would pay more attention to either making the software more tolerant of hardware/software variations, or identifying and making known which hardware/software can be "trusted" and which can't.

One other thing I've noticed that might be interesting to pursue: it seems to me that most (not *all* but most) of the posters on the forums who are happy with Sage's stability are not recording HD, and vice versa. Anyone else noticed this?

Eric

You are comparing hardware to software and the are different, so I'm not quite sure how your are countering my point about software. This is a fact of software development.

Perhaps you would understand if you wrote software instead of complaining about it. There are millions of variations of just software combinations not even factoring in hardware. Do you actually have a suggesting that can be implemented?

I would say that Jeff is doing a fantastic job considering the staggering array of variables that are in place. A very specialized processor in a TV is a world way from even a very general processor like an Intel CPU and CPUs have bugs. Remember the Intel floating point bug of the early 90s? And perhaps you've heard of GM? A couple of years ago, they recalled millions of cars. Phillips is recalling 12,000 plasma tvs.

Your point is inconsistent with reality.

Perhaps I am more tolerant because I actually write software for a living for 11 years and have a bit more understanding than the simplistic view being offered.

[Jesse]

Hi,

One other thing I've noticed that might be interesting to pursue: it seems to me that most (not *all* but most) of the posters on the forums who are happy with Sage's stability are not recording HD, and vice versa. Anyone else noticed this?

I have not paid enough attention to dispute that. But I will say that I am in the group that is happy and recording HD (and using a beta to boot).

I have only one issue. I am just gonna cave and stop using FSE. I am about 99% sure that will cure my issue.

Jesse

[aperry]

I too record HD (over Firewire) with few problems. Given that it is not "officially" supported, I'm not really blaming SageTV for the problems HD over Firewire recording causes.

[elaw]

Originally Posted by jominor

You are comparing hardware to software and the are different

Well duh! Only problem is I'm not doing what you're saying. Get yourself a blank 80C51 chip (pretty representative of what's in a TV set), hook it up and watch it do... nothing! Microcontrollers require code just like a PC does. And these microcontrollers are in things you use every day - TV's, microwaves, cars, VCR's. Ever seen one of those crash without provocation (no fair counting power spikes!)?

Originally Posted by jominor

I actually write software for a living for 11 years and have a bit more understanding than the simplistic view being offered.

That's nice... I've been doing it professionally since 1988. As a hobby since the late '70's. Right now I'm programming the control system for a multimillion-dollar deep space surveillance radar system, using PC's (XP machines in fact) and PLC's. And if it crashed 1/100 as often as your average Windows app, my employer would probably get sued.

Now I'm not trying to say it's sensible to hold Sage to the same standard. But I think just stating "all software has bugs" and leaving it at that is a cop-out.

[Opus4]

Originally Posted by elaw

Right now I'm programming the control system for a multimillion-dollar deep space surveillance radar system, using PC's (XP machines in fact) and PLC's. And if it crashed 1/100 as often as your average Windows app, my employer would probably get sued.

Will this product be launched into space? If so & it will be going anywhere near Mars, make sure one software team doesn't use metric units while another uses English units. I have no idea if anyone got sued over that one, though.

- Andy

[stevech]

hey - let's stop the debate about software engineering philosophy here. It's not useful.

[elaw]

Originally Posted by Opus4

Will this product be launched into space?

No, although I think many of the people working on it would like to see the whole thing go far, far away!

It's definitely not just software that has bugs, these poor guys can't even get their contractor to make the floor in the building flat!

[elaw]

Originally Posted by stevech

hey - let's stop the debate about software engineering philosophy here. It's not useful.

While I agree that maybe this has gotten a little too personal, I think software engineering philosophy is exactly the point of this thread.

It's all about that balancing act that applies to so many different things: between cost, functionality, and reliability. Personally I think Sage's cost is about right, but would rather see more reliability at the expense of functionality. But I know others feel differently and it's the Sage developers that are caught in the middle.

Eric

[blade]

Originally Posted by elaw

Well duh! Only problem is I'm not doing what you're saying. Get yourself a blank 80C51 chip (pretty representative of what's in a TV set), hook it up and watch it do... nothing! Microcontrollers require code just like a PC does. And these microcontrollers are in things you use every day - TV's, microwaves, cars, VCR's. Ever seen one of those crash without provocation (no fair counting power spikes!)?

How is comparing a PVR software program designed to run on literally millions of combinations of hardware comparable to software written to work specifically on any of the hardware you mentioned? All of the devices you mentioned perform very basic tasks and I'm sure the code is pretty simple. I know nothing about programming, but common sense tells me the more complex the function of the device the more complex the programming is going to be.

I'm sure if Sage set out to make the program as stable as possible on one specific hardware configuration and didn't bother to make it work with any other hardware they could design a rock solid system that never needed rebooting. The problem is everyone doesn't run the exact same configuration. No doubt Sage has some bugs in the software; however, trying to compare it to such simple devices seems a little unrealistic.

[aperry]

Originally Posted by blade

I'm sure if Sage set out to make the program as stable as possible on one specific hardware configuration and didn't bother to make it work with any other hardware they could design a rock solid system that never needed rebooting. The problem is everyone doesn't run the exact same configuration. No doubt Sage has some bugs in the software; however, trying to compare it to such simple devices seems a little unrealistic.

And this "new" all-in-one system they came up with could be called something like "Tivo"... Oh, wait...

[dblaine2]

Originally Posted by aperry

And this "new" all-in-one system they came up with could be called something like "Tivo"... Oh, wait...

Here! Here!

[roxy99]

Originally Posted by elaw

..but would rather see more reliability at the expense of functionality. But I know others feel differently and it's the Sage developers that are caught in the middle.

Eric

My demo version is about to run out. I would at least hope to have both reliability AND functionality. If that's not the case then maybe I should look elsewhere.

[Para]

I am very happy and satisfied with the reliability and functionality of SageTV. Is it perfect? No. Nothing is IMO. But it is a great piece of software and meets my needs nicely.

Would I like to see some changes or additions, sure. Who has not got their own preferences, wants or desires? But, I will not whine and complain about what the SageTV development team chooses to develop or how they prioritize. I'll take an oportunity every now and then to make some shameless suggestions and continue to help eveluate the betas. When you get the upgrades for FREE, it seems a little greedy to expect a whole lot more.

Shameless Suggestion number whatever...Trascoding ripped DVD's to the MVP Client

[tmiranda]

I've said it before, and I'll say it again. For the price Sage is very reliable. I reboot maybe once a month and everything works like clockwork. Could Sage be made 99.9999% reliable on all configs out there? Sure, it would just cost $100,000 a license or so.

I've spent over 20 years in software development and know that you can never get 100% reliability. But given enough money and time you can get close .....

[blade]

Originally Posted by roxy99

My demo version is about to run out. I would at least hope to have both reliability AND functionality. If that's not the case then maybe I should look elsewhere.

Keep in mind most people aren't going to post unless they have a problem. So when reading the forums you're going to read a lot from the few people who are having trouble, but you're not going to hear from the many more users who aren't having trouble. Any software you look at will have some users who complain about how unstable it is.

[src666]

Originally Posted by elaw

Well duh! Only problem is I'm not doing what you're saying. Get yourself a blank 80C51 chip (pretty representative of what's in a TV set), hook it up and watch it do... nothing! Microcontrollers require code just like a PC does. And these microcontrollers are in things you use every day - TV's, microwaves, cars, VCR's. Ever seen one of those crash without provocation (no fair counting power spikes!)?

Yes, and each of these chips interfaces with exactly ONE hardware configuration. They don't have to worry about what BIOS, CPU, Video Card, Capture Card, OS version, Driver version, Codecs, other software, etc., etc. the end user might have chosen to install on their system.

The level of complexity of writing bug free code on an open platform is infinitely greater than writing code on a closed hardware and software platform. Period. And as a developer, you know this already. So stop with the straw man arguments already.

[jominor]

Originally Posted by lotusvball

BTW, I am a software developer.

Then I eagerly await reading how you've acheived software perfection.