News From Gamepark Re: Gpkids/mini Development

[Alex.]

They were probably given suggestions on how best to market it to the GP32 crowd before release from people like Craig/ED though. Clearly a lot of them were wishful thinking.

That's the most likely truth, I agree. At least most of the 'promised' goodies were delivered by the community, so no harm done.

All GP2X software should run fine on XGP machines (with the exception of the GPkids, for fairly obvious CPU speed, RAM, resolution and OS reasons). They use compatible processors, compatible resolutions, will have sufficient RAM and graphical capability, and will apparently be using the same OS. Software won't be a problem.

You're talking about porting GP2X software over to the XGP. I was talking about native XGP software. Of course ports and transitions will happen, but I doubt that alone will be enough to give the XGP a solid foundation. Sure, you'll say that all the GP2X has now are emulators and ports, but there are already some great homebrew releases, and more will surely come after December 11.

you extolled the GP2X's praises ad nauseum, in much DaveC-esque fashion, at me at such a rate it must have made breathing difficult.

You're right, silly me not to agree with your anti-GP2X tirades. :rolleyes: How could I not have agreed that the stick is unusable, that AA battery slots are an insult to practicality, and that the build quality of the device makes potential publishers tremble in disgust?

- Alex

 

[Aninhumer]

Most of Epicenter's gripes are as a dev, not a user. Most of his bashing of the gp2x has been to do with the internals (Hard to use second core, No real 2D acceleration etc.).

I imagine this is why he cares so much about the specs of the XGP.

 

[Epicenter]

You're talking about porting GP2X software over to the XGP. I was talking about native XGP software. Of course ports and transitions will happen, but I doubt that alone will be enough to give the XGP a solid foundation. Sure, you'll say that all the GP2X has now are emulators and ports, but there are already some great homebrew releases, and more will surely come after December 11.

For me, if the XGP has all the software the GP2X does and is just built better (decent controls, build quality, battery-optimized design .. essentially fixing the GP2X's numerous faults) I'd be quite happy. There will be more unique software, but given the number of GP2X users who have stated they'd buy a new 2x unit that fixes the faults (which probably won't ever come) I imagine many would be quite thrilled with the XGP series.

Most of Epicenter's gripes are as a dev, not a user. Most of his bashing of the gp2x has been to do with the internals (Hard to use second core, No real 2D acceleration etc.).

I imagine this is why he cares so much about the specs of the XGP.

Bingo. GP's dedication to fixing any problems with the units before launch, which is quite different from GPH's approach, their willingness to meet the desires of the people who will be buying their systems. Gives me more confidence they'll be solid machines and good professional development platforms, at least much more so than the GP2X.

 

[Alex.]

So far GP has been all talk and no walk. At the end of the day, all you have is promises and a new product release delay. Just as you assume great things about GP, do the same about GPH, it won't hurt.

- Alex

 

[Epicenter]

GP doesn't have GPH's terrible track record, so I am not inclined to give them the benefit of the doubt. According to Craig/ED, they are still ironing out GPH's latest bad decisions as of this WEEK, so I am even LESS inclined. I'm not saying that I know GP will produce a great product, but I am at present expecting them to.

If all you're going to do in this forum is post 'GP's taking too long! They'll never release anything!' as DaveC and a couple others were doing-- please knock it off. You won't accomplish anything but start an argument. The idea of any discussion here is to share news and pool ideas to give to GP, as they want to hear our opinions regarding the systems' development.

 

[Shikaku]

4/6 AAs either as required or OPTIONAL. Hell, tell GP that they should also sell a AA battery pack holder or an li-ion battery seperately that will fit the XGP, meaning you don't have to buy the li-ion battery, but instead can opt for the AA holder pack, thereby reducing the cost. This can be only for the international sales obviously. Give us the option I say....

Please someone say that they have read this, at least the bolded.

 

[JaqMs]

I don't think having to spend $30 for high-capacity rechargeable batteries really reduces the cost. Anyway, having a huge battery compartment sticking out of the XGP/Mini would be quite unsightly.

 

[Shikaku]

I don't think having to spend $30 for high-capacity rechargeable batteries really reduces the cost. Anyway, having a huge battery compartment sticking out of the XGP/Mini would be quite unsightly.

The li-ion cell could be the same size as 2/4/6 AA batteries...

The DS Lite battery size IS about the size of 2 AA batteries in width and height, about half for thickness, just for your information.

It really depends. However, the point is OPTIONALITY, not AESTHETICS.

Also bear in mind that potentially 3 AA NIHM batteries couldlast more than twice as long as the original battery in the DS/DS Lite. Also bear in mind the XGP may be as thick as the DS Lite closed, meaning there might be plenty of room.

Everything is still up in the air, so chill. It's called a suggestion.

And even if they/you ignore me, I'll make a battery holder myself and sell it DaveC(tm) style if/when it comes out and if it is worth shifting.

 

[namco]

I prefer the S3C 2460 processor, it has more features (which means more software possibilities). The idea of choosing a lesser processor (even if they put on a sep graphics processor) could be a bad one.

 

[Epicenter]

Maybe GP is looking for ways they'd be able to upgrade the hardware for the XGP release, which would be a significant period of time after the XGP Mini, judging by this e-mail. The S3C2460 may be the highest-end chip they can really go for as far as Samsung is concerned, which would give them limited upgrade ability for the full XGP (more RAM, more NAND perhaps, but there'd already be a 3D accelerator, CPU, DSP, VFPU and video/image processor integrated that they couldn't upgrade further.) Texas Instruments makes some higher-end chips with ARM11s, but they appear to be very expensive and probably still will be when the XGP is ready to debut. Also I am sure that GP is getting some kind of better deal by buying more than 1 Samsung part, like Samsung LCDs alongside the SoCs to compensate the cost in buying a relatively small amount of parts.

If they did go with the approach of a lower-end SoC and then adding more features in the XGP than in the XGPmini, it would make sense, but it'd still be far from optimal. The 2410 is very much lacking in features-- they'd have to add a decent selection of coprocessors, methinks.

 

[Aninhumer]

I hadn't realised that the XGPmini was being released before the XGP (and in retrospect, this is quite sensible and obvious). I was intending to get only an XGP, but now I may end up getting both.

The XGPmini should have whichever option is best value (and not go over $150), if this creates a more obvious update path for the XGP, so much the better.

I terms of the upgrade from XGPmini to XGP, we don't know if Samsung will create new chips (unless someone does know they won't). By the time of the XGP release, they may have a new faster chip. It makes sense to wait a little before they start planning the XGP.

 

[shinneri]

So what's the logic behind making the mini and the XGP the same chipset at all? As long as they're both ARM and have similar specs (like both with 2D/3D video processor, etc), porting back and forth should be easy. After all, porting from the GP2X is supposed to be easy too (provided they will, indeed, run Linux).

I just don't know how GP will be able to justify 150 extra dollars for a widescreen, wifi, and DMB/GPS--especially if they plan to release it some time after the XGPmini.

 

[Epicenter]

I was thinking the same thing. Original assumptions abounded that the XGP/XGPmini would be coming out around the same time so it'd make sense if they had similar specifications as to be intercompatible as much as possible. Now that it appears there will be a sizable period of time between the two, there's not much problem with having 2 different SoCs for them.

The question is, how to go about differentiating the units. It seems doubtful to me that Samsung will have a very significantly more powerful SoC available, say, a year from now since they haven't released anything new in quite a while, but that when they DO it will probably be pretty expensive. I don't believe that GP intends to use SoCs from a company besides Samsung or MagicEyes since they are partnered. So the options would be to use the S3C2460 for the XGP and cripple the XGPmini in some way to reduce price (e.g. a lower-end S3C chip and perhaps just essential helper units), or to opt for a different company for the XGP's part and use the S3C2460 in the XGPmini.

It occurs to me that it's probably not unrealistic to decode/play video on a 266 MHz ARM9, so perhaps an MPEG4 decoder could be omitted. A relatively low-end 2D/3D accelerator might be called for, something that can handle extreme 2D work but isn't going to spit out 1 million triangles/sec.-- there can't be THAT much demand for extreme high-end 3D content on such a small display, after all. I do feel some sort of VFPU unit, like something AFP9-based, should be tacked on. Lack of an FPU cripples the GP2X pretty badly. Then the XGP would use the S3C2460 with the full 3D acceleration options and hardware MPEG-4 (it will, after all, be more work to decode a larger video frame for the resolution).

I will ask GP for clarification on the time between XGPmini/XGP release time and some hardware specifics, I will post a new topic when I recieve word on this.

 

[Aninhumer]

I think, if they intend to release the XGPmini earlier, it should have the best specs it can within the budget. There's no point crippling the XGPmini just to make the XGP better.

If, at the time for the XGP release, there really is no way to improve the specs (which I doubt), they can just stick in a widescreen, touchscreen, GPS etc. and sell it at an appropriate price, or even cancel the XGP.