SoC: Back and forth!


I still think it should be one GB ram. (3GB ram.)

Ram is cheap, right?
 
If OMAP 5's dev board can run 2 pcsx-rearmed and a lot of other things at the same time without slowing down, it's already more than I can ever ask for. 

What would you do with all that extra power? It's not like you can ever make it into a portable steambox. Sure, we may have saturn emulation or ran civilization a few month ahead of the android folks. It's simply not worth the risk. Even though Ti's not the best in the scene, we could at least count on it to not screw us over in the future. 

Thus, adding additional risks is not very smart in my honest opinion. Looking back to the ICP2 incident, it might turn out differently if backlit keyboard was not added. And of cause, the situation there could not be any worse if it was handled by pretty much anyone else or even by a chimpanzee with a death wish. 
 
For me, it's emulation.  I'm sure others would appreciate the added power too for that or whatever reason.
 
What would you do with all that extra power? It's not like you can ever make it into a portable steambox. Sure, we may have saturn emulation or ran civilization a few month ahead of the android folks. It's simply not worth the risk. Even though Ti's not the best in the scene, we could at least count on it to not screw us over in the future.
Honestly, we don't know enough about it yet to know if there is a risk to worry about. Wait until EvilDragon comes back with details before we talk about things being risky.

What we should do is look objectively at what it could bring us extra. This situation is very similar to the original Pandora where someone from TI contacted us through the forums and mentioned that he could get them into talks with TI for getting an OMAP. And you know what? That's what we ended up going with. We chose the OMAP because it opened up possibilities of N64 emulation, and would guarantee great PSX emulation.

Now we have a similar case here with Qualcomm. Whether we should do it or not will obviously be EvilDragon's decision, however, we should weigh the pros and cons of each, and look at what the Snapdragon chipset may open up for us. This will give EvilDragon a better idea of our position on this while he goes over the data they have given him to help him make his decision.

I would appreciate it if we could get some people who have a good understanding of the chipsets to offer their opinions on what would be feasible with this Snapdragon chipset.

We've seen Notaz mention some concerns he had based on past Snapdragon chipsets. I would be curious to know if any of those problems have since been solved or at least improved. I would also be interested in hearing the opinions of some other talented developers, such as Exophase, to hear their thoughts on the Snapdragon chipset.

-God Ginrai
 
@ekianjo

How's it going with the Snapdragon 800 discussion with Qualcomm, is ED talking to the appropriate contact now?
We had a first positive feedback, now we are waiting for further news. Give it a couple of days. 
Any chance for the 805.

Don't know if the 800 is the best choice.

Would definitly go for the 805.
 
-snip- 

I would appreciate it if we could get some people who have a good understanding of the chipsets to offer their opinions on what would be feasible with this Snapdragon chipset.


-snip-
So I'm none of the people that you mentioned, but I can shed some light on the chipsets and emulation. If you want to see how various chipsets perform on various emulators then user reported statistics should be fairly available online. My Nexus 5 plays everything through N64 just fine. Haven't tried Dreamcast. PPSSPP runs many games at nearly native speed, and that's on an older revision. I've worked on various personal emulators in the past and know enough about them to make at least a somewhat educated response about how hardware will affect emulation capability.

Changing between OMAP5 and Snapdragon800 (for example) won't magically enable PS2/Gamecube/Xbox emulation or super fast windows XP emulation. The main problem with emulation isn't how fast our hardware is; it's whether an emulator for a given system exists. Take PCSX2 as an example. Sure, PCSX2 exists, but an ARM port doesn't, and that project seems to be on its last leg. Porting PCSX2 to ARM is not trivial at all, since it will require an incredible rewrite of nearly all of PCSX2. Hardware capability is not the limiting factor in what the Pyra is capable of (at least regarding emulation), software is. Emulation is hard. Really really hard. If it was easy then PCSX2 would already be running on ARM. Which it isn't.

The main stumbling block with Qualcomm chipsets is the whole "open" concept that the Pyra is aiming for. Qualcomm is not one to give up its intellectual property easily, and I think this is one of the reasons why publicly available documentation is sparse and open source support is shoddy. Sure, there are open source projects that are helping to support Qualcomm devices (like Rob Clark's freedreno project), but a successful device won't exist without tons of support direct from Qualcomm.  

I figure I should also mention that I'm a huge Qualcomm fanboy. So much so that I interned for them last summer and will be doing so again this summer. Still, I know the company isn't perfect and that I'm not sure that Snapdragon will help the Pyra achieve the "openness" that it wants to. 
 
What we should do is look objectively at what it could bring us extra. -God Ginrai
We can weigh this From both prespectives. On one hand, end users like us would want the Pyra to be as good as it can be. So we think about things like competability, power, Opengl support, battery life and so on.

On another hand, ED has another set of completely different factors to consider, such as: market margin, profit margin, difficulty in production, keeping the core community members and so on. Qualcomm's socs are rather expensive, and drain power like no others. Most us would buy the pyra regardless of wether it's using the OMAP or the Qualcomm. Pyra would cost more to produce not only for the soc, but also potentially for a larger battery. (I'm not sure if this is the case for the 805) This would be a pure loss to ED as it will reduce ED's profit margin if ED does not rise the price tag. If it's more expensive then less people could afford it. Since many of us had expressed their interest in pandora's DSP, switching to Qualcomm could be a turning down. 

I say, we should consider Qualcomm as an alternative, but we should also make sure to avoid most risks so the pyra can be successfully pushed into the market. ED still needs to pay back his investors, and he's not in a particularly good point financially. (ven though he's optimistic about it, he doesn't have the capital for mass production. Anything can happen. In the worst cases, he could be left with no money to further his business, and would have to abort the project. (it can be avoided however, unlike in Craigix's case. ED's pretty capable on this sort of stuff, but we should not count on his luck forever.)   :blink:
 
Last edited by a moderator:
I don't care about ARM or x86 or MIPS or whatever, I don't code in assembler (or at least not enough to care about it). As long as there's a FOSS compiler available, I'm happy.

Binary compatibility is not important to me -- the point of FOSS is that you can recompile easily.

Availability (in low quantities, now and in a few years), suitability to the form factor (board space, thermal budget), performance and perf/W, having FOSS drivers available, and maybe price. Those are important to me, roughly in that order. Others may have different priorities.

I think the OMAP 5, SnapDragon 8xx, and Z3770 are probably all good choices, at least on paper.

One "sociological" disadvantage of x86 is that it is likely to attract hordes of gamers who are attracted to the idea of a handheld that can run proprietary x86 PC games. It would most likely dilute our community to the point that the proportion of nice, smart and interesting people is reduced to homeopathic levels.

Other than that, I have no problem with x86 (if it scores well on my criteria, which seems to be the case on first sight). But if ED and notaz want to stay with ARM for whatever reason, that is fine with me too -- they are the ones who are actually making the Pyra, it's their decision.

I must admit that the improvements in power consumption Intel made are impressive -- although it's a bit hard to compare SDP with TDP etc. But this is the first generation of Intel SoCs that finally has a perf/W that is competitive with ARM. It makes sense to wait and see if they are really that good. The Z3770 is designed for tablets -- those have a lot more room for battery and heatsink than we have. How does the TDP of the Z3770, OMAP 5432 and SnapDragon 805 compare?
 
One "sociological" disadvantage of x86 is that it is likely to attract hordes of gamers who are attracted to the idea of a handheld that can run proprietary x86 PC games. It would most likely dilute our community to the point that the proportion of nice, smart and interesting people is reduced to homeopathic levels.
Yes, thus making is much less interesting to hang out here. I don't want these boards to become another reddit.   
 
The z3770 has open source drivers.  It is supported  in the Linux kernel.  It meets 100% of the other requirements for an SoC for this system, but it is X86 not ARM.
 
It is even available though Mouser in single unit quantities at around 40 EUR. They even have 100 in stock. But no data sheets :( I.e. we can't even check if and how we can connect anything.


What makes me nervous is that it is said to be a "FCBGA1380" in a 17x17mm package (OMAP5 is 17x17 mm PBGA754). This means it may need a more expensive pcb technology to be connected. And if I interpret some rough diagram correctly, it has only one eMMC and one SD card interface. This makes interfacing two SD slots plus internal eMMC more difficult (could go through USB and an extra chip) and WiFi as well. So the total system complexity must be taken into account.


But it is quite useless to buy chips without getting public documentation. For the OMAP5 there are 6000 pages of technical reference manual available.


At the moment it looks to be the first time that we have a small chance to get an up-to-date SoC with x86 architecture (it was impossible for the last 10 years). But if this is the right train to board is a different question.
Data sheet is linked to from the Intel product page:

http://ark.intel.com/products/76760/Intel-Atom-Processor-Z3770-2M-Cache-up-to-2_39-GHz


Right hand column. "Download Datasheet" It's not like they hid it.


I have no basis to argue whether the PCB technology is more or less expensive.


eMMC and SD interface overview is in section 1.2.11 of the datasheet.


Supports one SDIO 3.0 controller


Supports one eMMC 4.51 controller


Supports one SDXC controller


Personally I'd put the WiFi on a USB 2.0 port so that it's easier to swap out in the process if the WiFi chip chosed goes EOL. Isn't that easier and would have faster data?


Per page 21 of the datasheet, the Z3770 has One USB 3.0 OTG port and an internal hub with 4 USB 1.1 to 2.0 ports.


Anyway, the documentation doesn't get any more 'public' than posted in the open on the product page. There are 291 pages on V1 (overview) and there is a 23MB V2 as well - I'm pretty sure you'll find anything you need there IF the project is serious about considering this option.


I agree that it is possible that the Z3770 may not the right SoC for this project. However, Intel has seemingly pulled a huge rabit out of it's hat and seemingly created exactly what our specs call for.


It's fast, efficient, available, has open source drivers and is relatively innexpensive.


It may not be for us, but wow - we should at least consider it.


Of course there is something attractive to me about having a 64bit X86 Debian handheld running Half Life and Neverwinter Nights. So I'm biased.


Before someone piles on the 'We've been through this! It's Done! You're an idiot for bringing it up!'


I was directly addressed from the system designer with their concerns - and misconceptions surrounding this chip that they apparently had -not- considered. This reply is to allow them the option, if they choose to explore it, to see the information that was claimed to not exist.
this is indeed the missing piece of information. Maybe it is not that prominently published as by TI or simply not that well known (meaning intensively linked to from the outside).

Many thanks for the link to the data sheets. They indeed look impressive.

Well, we *may* be able to offer different and exchangeable CPU boards if we can squeeze them all onto a small and flat CPU only board... Maybe in a second step... Just dreaming :)

Anyways it is EDs decision.
 
Last edited by a moderator:
Well, we *may* be able to offer different CPU boards if we can squeeze them all onto a small and flat CPU only board... Maybe in a second step... Just dreaming :)
That would actually be a bad idea. Offering different architectures would fracture the community and development for the device.

EDIT: One thing worth noting is that regardless of PNDs being able to "just work" on the Pyra or not, sticking with the ARM architecture allows developers working on a game on the Pyra to release it for the Pandora with minimal changes required and vice versa.

-God Ginrai
 
Last edited by a moderator:
I must admit that the improvements in power consumption Intel made are impressive -- although it's a bit hard to compare SDP with TDP etc. But this is the first generation of Intel SoCs that finally has a perf/W that is competitive with ARM. It makes sense to wait and see if they are really that good. The Z3770 is designed for tablets -- those have a lot more room for battery and heatsink than we have. How does the TDP of the Z3770, OMAP 5432 and SnapDragon 805 compare?
Actually, it is the second generation that is competitive in terms of performance per watt. The previous generation (32 nm) was already pretty good when compared to A15 based SoC. Anandtech did a pretty detailed comparision ( http://www.anandtech.com/show/6536/arm-vs-x86-the-real-showdown ). The new generation is supposed to be even more power-efficient. Cortex-A15 based SoC are usually also designed for tablets, because of their fairly large TDP.

EDIT: One thing worth noting is that regardless of PNDs being able to "just work" on the Pyra or not, sticking with the ARM architecture allows developers working on a game on the Pyra to release it for the Pandora with minimal changes required and vice versa.
For games this should rarely matter, unless you are coding it (or parts of it) in ARM assembly. This is usually not the case.
 
The architecture shouldn't be the deciding factor. There are x86 fanboys and thare are ARM fanboys. Like wej said, it mostly doesn't matter what architecture you're compiling for.

I very much like Intel as a company. They are super Linux friendly and they are on third place of code contributions to the Linux kernel, followed by TI and Linaro. Their graphics drivers are the best in the Linux world. On top of that they are open source! So you can write changes and patch stuff. That would make it possible to have a completly open source Pyra OS without sacrificeing power or capability! - pretty cool reasons to consider if you ask me ; )

Anybody thought about Steam yet? They have a lot of really nice indie games. I'd love to play them on the Pyra with the sweet Nexus5 screen *drooling* ; )

Edit: this thread is like crack ha ha ha! xD
 
Last edited by a moderator:
^ quotation from the article: "They might have a backdoor that allows spooks to access and control computers." - they might have dead aliens in area51 : p

anyway... *rofl*... they don't need a backdoor. Maintenance and debug functions are scary enough xD

https://en.wikipedia.org/wiki/Intelligent_Platform_Management_Interface

How about a donation run to get gta04 an Intal dev board for futher testing and comparison?  ; )
 
Last edited by a moderator:
Back
Top