I was just gonna say that! I wanna see its insides. (We're cruel and callous).
Digital Foundry tech analysis channel at Eurogamer
- Thread starter FirewalkR
- Start date
- Status
If you're getting a PS3 Slim, you may also be getting the 3.0 Firmware. If so, that would of course be interesting to look at. If it comes with the LBP theme, you could do a timelapse on that also 
(not really important though).
The hardware side new bits should be:
- being able to output sound to multiple outputs at the same time
- Bravia unified support (standby for TV and PS3 with one click, shared menus I think also?)
- As per the above, the HDMI port should now be bidirectional? At any rate, something has changed there, so you may want to do some simple captures and see if the default output is different at all
- harddrive slot now in the front (though I think we've got videos online, but it's a popular thing)
- is it quieter, louder or the same?
- there's probably a different BD drive in there. Not sure how you can check that though - maybe it sounds different?
- buttons have changed
- different power-plug (like the PS2 again and unlike a PC) but still no brick as far as I can tell
- it still stands on its side pretty well even without stands
- it weighs 3,5kg, the original was what, 6kg?
- matte finish (maybe run your fingers over it and show the difference in visibility of grease left behind )
- it should be able to output bitstream as well as lpcm this time
That's all I remember from the top of my head about the hardware changes (quite a lot of changes actually when you add them up)
(not really important though). The hardware side new bits should be:
- being able to output sound to multiple outputs at the same time
- Bravia unified support (standby for TV and PS3 with one click, shared menus I think also?)
- As per the above, the HDMI port should now be bidirectional? At any rate, something has changed there, so you may want to do some simple captures and see if the default output is different at all
- harddrive slot now in the front (though I think we've got videos online, but it's a popular thing)
- is it quieter, louder or the same?
- there's probably a different BD drive in there. Not sure how you can check that though - maybe it sounds different?
- buttons have changed
- different power-plug (like the PS2 again and unlike a PC) but still no brick as far as I can tell
- it still stands on its side pretty well even without stands
- it weighs 3,5kg, the original was what, 6kg?
- matte finish (maybe run your fingers over it and show the difference in visibility of grease left behind
)- it should be able to output bitstream as well as lpcm this time
That's all I remember from the top of my head about the hardware changes (quite a lot of changes actually when you add them up)
grandmaster
Veteran
Won't many if those changes apply to the fat PS3 too once 3.0 is installed?
And yes, I hope to dismantle it.
And yes, I hope to dismantle it.
Nope, I only listed the things which I think are hardware specific to the Slim.
Well, I'm not exactly sure, but I think it has something to do with the fact that the PS3 currently can only send CEC information but not receive it, or something like that. I'm not sure about this, and it may still be a firmware thing, but this is what seemed to be suggested in one article I read, without being too specific.
Why would HDMI need to be bidirectional to control Bravia with a controller? It's no different to using a remote. The PS3 doesn't need any feedback; it only has to send non-video data down the HDMI. Unless that's what is meant my 'bidrectional', instead meaning something more like 'dual-channel'.
Isn't it the other way around? You can control the XMB with your Bravia remote?
Also, according to Gizmodo the Slim does not come with Firmware 3.0. http://gizmodo.com/5341207/10-things-you-need-to-know-about-the-ps3-slim
Duuuuh! How dumb am I?!
Oh yeah this reminds me, since we have different harddrives and you're opening up the slim, it could be interesting to see if the different harddrives have a noticeable performance influence on games. This could be a separate feature of course.
grandmaster, what I'm curious about is whether there is a measurable difference between various PS3s (slim or otherwise) especially when it comes to CPU tasks, given that one of the SPEs is randomly disabled. I think a game like Sacred 2, which is apparently CPU bottlenecked may prove to be a nice benchmark.
I don't think that's going to work ... there aren't enough games that use the SPE's enough concurrently with enough data going. In the very rare case that there would be an application that would flood the EIB enough, using enough concurrent jobs on enough SPEs to be able to notice adjacency, I'm almost 100% sure they would write a tiny bit of code allowing them to identify which is which and adjust the code accordingly.
You don't need to flood EIB, you can already measure EIB latency of SPEs from PPE with minimal communication. I think you only need a game that's sensitive to that latency (which may not be the case if CPU boundness source from not having time to move PPE code to SPEs).
As for the auto detection of SPE adjacency, I'd be impressed if anyone bothers with that, because I don't think the difference will be enough to care especially with any robust code. Plus I believe SPURS jobs are scheduled automatically without knowing the needs of the code.
As for the auto detection of SPE adjacency, I'd be impressed if anyone bothers with that, because I don't think the difference will be enough to care especially with any robust code. Plus I believe SPURS jobs are scheduled automatically without knowing the needs of the code.
grandmaster
Veteran
Surely the games would simply be coded for six SPUs and it wouldn't matter which of the eight were disabled?
I'm always thinking that it's disable for all PS3… And if no I don't think that games are design to use a "hypothetic" SPE… Sony more keep it for OS IMHO.
Let's do a refresh here.
There are 8 SPUs total, and one of them is randomly disabled to increase yield. (Even if that's not the case now, it's still not accessible by games.)
Among the remaining 7, one is completely, one is partially reserved by OS.
So all games have access to 5.5-6 SPUs.
Now EIB latencies obviously depends on which SPU is disabled (and which one is reserved by OS), because that will change the locations of the ones you have access to. And this is easily measurable on Cell.
The only question is are there games out there that are sensitive to that latency to the extend video output is measurable different.
There are 8 SPUs total, and one of them is randomly disabled to increase yield. (Even if that's not the case now, it's still not accessible by games.)
Among the remaining 7, one is completely, one is partially reserved by OS.
So all games have access to 5.5-6 SPUs.
Now EIB latencies obviously depends on which SPU is disabled (and which one is reserved by OS), because that will change the locations of the ones you have access to. And this is easily measurable on Cell.
The only question is are there games out there that are sensitive to that latency to the extend video output is measurable different.
grandmaster
Veteran
OK I see where you are coming from now. The issue is that if you run the exact same video through any console twice (for example, a stored replay of Ridge Racer 7) you will always get a very small variation in the results. This could be down to any number of factors - for example, background processes related to the OS. So it would be very difficult to measure and say for sure it would be down to SPU latencies.
- Status
Similar threads
