DonFerrari on 03 May 2018
Pemalite said:
potato_hamster said:
Care to explain why Nintendo went with a modified Tegra 1 instead of a Tegra 2 variant in the Switch?
|
Same reason why they chose the technically inferior (Older+Slower) chips in the Wii and Wii U probably?
I cannot ascertain why Nintendo chose the chips it did, but I can tell you that it wasn't the latest and greatest of chips on release.
It's also not a modified Tegra 1.
It's a plain jane Tegra X1. The die shot between the Switch's SoC and the Tegra X1 was identical.
potato_hamster said:
Care to break down how much say a Vega. or Pascal / Ryzen based APU would cost compared to the current cost of the PS4's APU?
|
That is entirely dependent on yields, size of chip, how high they clock, how much voltages they push through.
Obviously I am not the person who is fabricating and packaging these chips, thus I am unable to give any quantifiable cost numbers empirically.
With that in mind, you only need to take a look at the current Ryzen APU's on the market to get an idea of costings.
potato_hamster said:
Care to mention how much "4x 18650's" cost vs the Switch's battery, how the size of the batteries compare, and if these 18650s require any additional shielding or protection circuitry to prevent portable PS4 portables from maiming their users if something goes awry?
|
The packs I have built... Probably about $20 tops.
And the packs I have built have 72x 18650's, you do need a BMS and so on, but when you are working with only a few cells it doesn't need to be overtly complex or expensive.
But I should also state that current devices like the Switch already have a BMS and Battery... So it's not going to be a real increase in cost on that front.
potato_hamster said:
No one's really questioning whether the technology is there. There's $1200 laptops less than an inch thick with X86 processors, and AMD GPU solutions that will eat the PS4's lunch and go dead within an hour. No one is disputing that. But now we need to make such technology less than half the size, less than half the price, and double the battery life. So please, sort out the feasibility of making such a solution less than say, $400.
|
The thing with Laptops however is that OEM's like to have their profit margins, because they are a business who is trying to make their cash off these devices alone.
Console manufacturers tend to go with lower/non-existent profit margins and make up for it with software.
Then you have bulk-purchasing.
In short... Looking at a laptop and the hardware it may/may not have is not representative of the potential of a gaming handheld.
There is also a significant increase in "baggage" in a laptop compared to something like a tablet... Like the software stack, you don't need a $50-$100 Windows OS on the Playstation 4 Portable do you?
zorg1000 said:
Is there a reason why you are ignoring price? The person you responded to has repeatedly made it clear they arent just talking about whether or not such a device is possible but also if it can be released at a reasonable price.
|
Possibly because I can only go by what is happening in the general marketplace?
Besides... This entire thread is speaking in hypotheticals, which is why my stance is: You cannot gauge the pricing of a Playstation 4 Portable, but certain aspects (Battery and so on) can be improved over the Switch for minimal cost by looking at what is available on the Market.
I mean Shit. Even nVidia has better Tegra SoC's than the Switch and had them even when the Switch release.
|
How dare you say Nintendo is using a solution that was basically about ready? They are the sole inovators of the industry with everyone else playing catch up... also please ignore all other hybrids or the HH that could be played on the TV, they all don't count because only Nintendo made it right.
twintail said:
Pemalite said:
Sure.
I don't think people realize how inefficient the Playstation 4 SoC is. - You don't actually need a chip that is identical to the PS4 chip to achieve the same outcome.
Ergo. You don't actually need a 1.8 Teraflop chip.
The Playstation 4 chip was outclassed even on it's release, all of these years haven't exactly been kind to the Graphics Core Next 1.0 architecture or those based upon it. It's old, it's slow, it's inefficient, it's hot, it's power hungry... Vega and Polaris are significantly more energy efficient, but still pale in comparison to say Maxwell or Pascal, let alone Volta.
CPU wise, Ryzen slaps Jaguar every day of the week, throw it down on 12nm, get aggressive with binning, consolidate some other parts of the system into the SoC, lower clocks and voltages... And bobs your uncle.
And then get aggressive with battery chemistry, LiCoO2 or LiMn2O4 are good candidates or maybe even LiNiMnCoO2.
The Switch's battery is good for 4,300mAh.
4x 18650's could provide 12,000mAh.
|
I appreciate the reply and the indication of viable tech points. I still dont see all of this means a reasonable price point anytime soon.
| DonFerrari said:
Well you know there would be people that would buy a PS4 portable with the same games just because it's portable. We have a lot of people in VGC that put games having twice the value because you can play HH and console, or that a game is made better simply by being portable, etc... so there would be some market for a portable PS4.
|
Dont get me wrong, I am sure that market exists. I just doubt its size to be sustainable enough.
|
The size is totally arguable and if you said it is small i don't think any of us could prove the size anyway =] but perhaps they could get additional 30M users, but all things considered I don't think the impact would be to expressive for the cost.
