[POLL] Snap Dragon vs Nvidia Tegra as the soc.

what should the soc be.

  • Snap Dragon

    Votes: 10 40.0%
  • Nvidia Tegra

    Votes: 3 12.0%
  • other

    Votes: 12 48.0%

  • Total voters
    25

dfist45

Still Fresh
Joined
Dec 23, 2012
Messages
9
discuss both chips and why you would chose one over the other for the next Pandora/thing/whatever it will be called.

or do you have a different idea
 
ED is already sourcing a chip, but has not said what it will be yet. Tegra is very unlikely, because it would be very difficult to get in the low quantities required. Whichever SoC has the best compatibility with the current Pandora will likely be chosen, so again, probably not the Tegra.
 
Out of those two, I think Snapdragon is the best choice and maybe also the more realistic one.

Why did you only consider those two? What about OMAP5, Exynos, and all the chinese stuff?
 
ED has previously failed to get ahold of Exynos.

I suspect it'll be an OMAP, as ED has been in contact with TI before about it. Or possibly a FreeScale i.mx6, which pretty much any schmuck with a credit card can get in quantities of one.
 
Last edited by a moderator:
I thought the critical factor was MOC, not which of the many chips out there is the most suitable...
 
I hope it will be the Omap5, it has the same DSP so the DSP code will work too. And the DSP is very very mighty.

http://www.isee.biz/products/igep-processor-boards/igepv5-omap5432

Edit: the only problem I see is that the PS2 emulator could use 3 cores, and it is only a dualcore cpu. But one (or maybe even both) of the vector engines could maybe handled by the dsp, so this would be a big speed gain there.
 
Last edited by a moderator:
I simply voted for Snap Dragon because of the name. :D  Tegra would be nice but I doubt it's available for small batches.

I don't like TI's OMAP, to much trouble with that in the past, see Pandora and all the issues (poor support from TI, driver issues, hard to use DSP and so on). 
 
Last edited by a moderator:
I hope it will be the Omap5, it has the same DSP so the DSP code will work too. And the DSP is very very mighty.

http://www.isee.biz/products/igep-processor-boards/igepv5-omap5432

Edit: the only problem I see is that the PS2 emulator could use 3 cores, and it is only a dualcore cpu. But one (or maybe even both) of the vector engines could maybe handled by the dsp, so this would be a big speed gain there.
According to Wikipedia the OMAP5 has 2 Main cores and 2 sub cores.

Also it has a DSP and NEON.

There should be a way to replace that 3rd core ;) .
 
Wikipedia says: 


Because PCSX2 only uses two main threads, having a CPU with more than two cores does not significantly increase PCSX2 performance. Performance can be improved up to 70% by enabling the multi-threaded VU (MTVU) hack in the Speedhacks menu, which will use an additional third thread (thus mostly useful for CPUs with more than 2 threads).
That said, maybe pulling input, audio, etc onto a DSP/little core could help.
 
Let's use a SoC that allow us to have better perf than current pandora, keep backwards compatibility and low power requirements so that we keep 10 hours off the battery with pandora softwares.

TBH, I don't care about the brand…
 
TBH, I don't care about the brand…
I don't care about the brand too, but I care about DSP programms. I am not sure if other SoC companies than TI will include this specific TI DSP on their soc. I think they won't.
 
Last edited by a moderator:
According to Wikipedia the OMAP5 has 2 Main cores and 2 sub cores.
Also it has a DSP and NEON.

There should be a way to replace that 3rd core ;) .
Everything ARM that would be a viable option has NEON. It's not a separate processor, it's a part of the CPU cores.

It'd be good if people at least waited until the DSP on Pandora was used for something substantial before hyping C64x on Pandora 2.. On OMAP4 they gimped the DSP (in other words, it's weaker than it is on Pandora), unless they substantially de-gimped it for OMAP5 it's going to be pretty piddly compared to the CPU cores. Based on TI's promotional material I don't think they de-gimped it at all:

"A programmable multimedia engine based on TI’s mini-C64x DSP and power-efficient, multi-format hardware accelerators"

The thing is really just there to carry over media playback.
 
Last edited by a moderator:
i was hoping some kind of word what Exophase would hope for :/ i'd still choose omap5 over these other two, not because i want omap, but because i dont want the other two
 
It'd be good if people at least waited until the DSP on Pandora was used for something substantial before hyping C64x on Pandora 2..
There already is a sprite/blitter engine for the Pandora DSP that (now) is (almost exactly) twice as fast as the GPU and faster than any ARM/NEON blitter but apparently there is little interest in using it (a tweaked version of SDL could help with that. Any volunteers?)

(..I'm also working on some things which look really promising so far..)


The DSP on the OMAP5, according to specs, does not seem to be an improvement over the one in the Pandora. It _may_ even be slower since TI removed some performance-critical features (only 4 execution units instead of 8, no software pipelined loops, and the L1DSRAM can not be memory mapped any longer (but L2 still can..and there's this L1 cache maintenance/lockdown feature..)). Last but not least, it's not binary compatible to the OMAP3 DSP.


However, if the "P2" will be based on OMAP5, I'm definitely on board. The DSP may not be the best choice for 2D graphics, anymore, but that does not mean that it will be useless. Plus there are two A15 processors, one of them could always take on the role of the Pandora DSP (maybe even at approximately the same power consumption level).


The main benefits of a "P2" will most likely be GPU speed(!), increased resolution, and improved connectivity (HDMI/USB3_or_SATA/..).


Sticking to OMAP5 definitely makes sense since the scarce low level/kernel developer resources in this community (*cough*, Notaz, *cough*) are already familiar with the OMAP family, and it retains important HW features (e.g. display controller overlays / video scaler).

p.s.: regarding "hard to use DSP and so on": Funny you should say that because in reality things are quite different: ARM/NEON is hard to use (read what Exophase wrote about this topic! i.e. NEON is poorly documented and manual cycle-counting and assembly programming is mandatory) while, in my experience, compiling standard ANSI-C sources with the TI DSP compiler already results in very good performance that is mostly limited by the available memory bandwidth.
 
Last edited by a moderator:
If there's a sprite blitter out for the DSP then it's no wonder anyone is using it, I can't find a thread for it anywhere and even searching for posts isn't bringing up anything..

I'm not specifically against the Pandora successor using OMAP5, I'm just reacting to people who are acting like it'll be ahead of other SoCs because it has a vaguely OMAP35xx-like DSP. As far as CPU perf and perf/W and GPU capability go it's not particularly special. But if no other SoCs are a viable option then how it compares is moot.
 
discuss both chips and why you would chose one over the other for the next Pandora/thing/whatever it will be called.

or do you have a different idea
Whats the use of this thread ? Both options are not gonna happen due to availability to Ed, so why bother on deciding which we should mourn for ?
 
There already is a sprite/blitter engine for the Pandora DSP that (now) is (almost exactly) twice as fast as the GPU and faster than any ARM/NEON blitter but apparently there is little interest in using it (a tweaked version of SDL could help with that. Any volunteers?)
Do you have a link for it?
 
I'm still holding out hope for the z3770.

From everything I can find, it's simply a much better chip for our type of device.
 
The funny thing is, by the time the P2 comes out, neither of those two SOCs will be current.  At minimum , Tegra 5 will be out etc.

I wonder if ED could get his hands on dev samples of next gen SOCs, I did hear something about Nvidia sending out Tegra 5 dev units to some manufacturers, could be wrong. Although , I'm not sure what luck, if any, ED would have of getting one. May just have to live with being 1-2 gens behind with the SOC
 
Last edited by a moderator:
Back
Top