It's not x1, this isn't the retail switch, it's something else if real. So those clocks don't apply.
Honestly, it says 516 cores, I just don't see it. I'm keeping expectations super low, like lower than a cheap stripper low.
It's not x1, this isn't the retail switch, it's something else if real. So those clocks don't apply.
Is it the same people? Do you have any examples?
Honestly, it says 516 cores, I just don't see it. I'm keeping expectations super low, like lower than a cheap stripper low.
Just out of interest, why do people repeatedly say they don't want a powerhouse console from Nintendo, that they would prefer they did their own thing and then spend many, many pages trying to work out if Nintendo have some hidden hardware, or if their hardware is actually more powerful than it first seems?
They did the same for the Wii U and now the Switch, I thought hardware power didn't matter?
So it supports what is shown here?
Dat L-shaped heat pipe. FoxConn leak strikes again!
[Edit: Nvidia weren't kidding about how much work went into this thing to make it possible, from all facets. Just looking at the dense, yet clean and thoughtful layout of the board is telling. It makes certain comments about the system hardware kinda lame, as though it's a trivial thing to add more RAM, Battery capacity, storage, etc, in a space this tiny. It's a real disservice to engineers who devoted that much time and effort. I'm impressed, and hope for a round-table event similar to "Iwata Ask" special that detailed Wii U design]
Cuningas de Häme;230645050 said:Broken Hope is talking about the straw men hiding under his bed...
What is wrong with some people? I mean, most of the folk in this thread has been wonderfully active in the discussion, trying to figure out what's inside the Switch. Then we have some coming here, yelling about people talking about better than PS4Pro and how delusional people are...
Then we have the one who just loves Ls.
There has been absolutely no single person that has said that the Switch will be even on par with XBOne. They are speaking about what the Switch can theoretically achieve, as in best case scenario. They are not saying it will be that, in fact most of them add that they believe it's something little less powerful than the aforementioned best case. Why people try to derail a wonderful tech thread spouting nonsense about things no one has said?
Chinese speaking guys, the usual Switch thread suspects and other sensible folk who actually try to engage in discussion... You're great! Keep on trucking!
Deeke[VRZ];230627538 said:If this is the RAM code:
K4F6E30
4HBMGCH
Then the "F" in K4F means "FP" or "FPM" which translates to a variety of RAM called Fast Page Memory RAM.
https://gyazo.com/6591bbbecf2542c41515f9ba310b8b0b
If this is accurate, I think the two modules have different codes.
I'm not 100% sure if it's viable for both modules to be FPRAM or not. Still decoding!
Deeke[VRZ];230634322 said:RE: Switch RAM
I think the K4F6E304HB-MGCH code is right. Searched the web and found this.
http://www.alldatasheet.com/view.jsp?Searchword=K4F6E304HBMGCH
AVNet distributes the RAM, and Samsung Electronics (SEC) makes it.
According to this link, K4F6E304HB-MGCH modules are 16Gb DDR4 modules (Gb = gigabit)
link: https://products.avnet.com/shop/en/...56?CMP=EMEA_FindChips_inventoryfeed_VSE&c=EUR
16 GIGABIT = 2 GIGABYTE
DRAM Chip Mobile LPDDR4 DRAM 16G-Bit 512Mx32 1.8V 200-Pin FBGA
Here's the datasheet.
http://www.samsung.com/semiconductor/products/dram/mobile-dram/K4F6E304HB-MGCH?ia=3107
EDIT: I'm told 3Gb/s isn't that fast. But it's LPDDR4 which is more efficient than previous RAM.
Deeke[VRZ];230636066 said:More Switch RAM info:
I couldn't find exact frequencies for the K4F6E304HB-MGCH LPDDR4 modules, but I did find speeds for a K4F6E304HA-MGCH module, which are in 2GB densities.
Density 2GB
Organization x32 (2CS,2CKE)
Speed(Mbps) 1600MHz
Package 200-FBGA-10.0X15.0
Temperature(C) none
Production Status Preparation
These modules could be very much alike.They are built on the same 200-FBGA platform and have the same 16Gb (2GB) densities.
In fact, I'm thinking the "B" series is a derivative.
https://memorylink.samsung.com/ecom...4HA-MGCH&partSetNo=LPDDR4&partSetLabel=LPDDR4
Begs the question, why this small? They could have made a bigger tablet
Begs the question, why this small? They could have made a bigger tablet
Just to get this out of the way before I'm accused of being a fanboy, I have the Switch pre-ordered and am getting it at launch, the first Nintendo console in years I've actually done that.
Nintendo have actually done a fair bit right with the console. It no longer looks like it was designed by Fisher Price, it has a screen that doesn't look like dog shit and it looks like they finally sorted out their account system.
But having said that, why aren't people allowed to criticise stuff without being jumped on?
Why is their a subset of Nintendo fans that will back up and defend every decision no matter now stupid?
As said before, if it was underpowered that's fine because art style is more important, if it was PS4 Pro level hardware that would suddenly be amazing too because only power matters.
Then you will have the people that are obsessed that the Switch is more powerful than it is and use games like ARMS to back that up, when the only reason it looks decent is it has good IQ and decent textures due to the system having more RAM.
I think this detective work needs more appreciation.
If I read this right, this would mean a 64bit memory bus, right?
I'm not talking about English though. I'm talking about alphabet/writing. I'm sure you're aware about the existence of different writing, yeah?
A writing in which "L" doesn't exist.
Yes, this confirms that a prototype at the end of july had 4GB of 64bit memory. Which is also likely the july devkit which we already have documentation for.
I'd refer to anything that is basically a right angled item as 'L' shaped regardless of what orientation you actually deploy it in. L literally draws the shape and you instantly know what is being talked about. wtf is 'J' shaped - you'd just confuse someone talking about that.
Begs the question, why this small? They could have made a bigger tablet
We don't know if it's a prototype or if it was manufactured at the end of July.
What we know is that the SoC was manufactured in July.
And we have a Reddit post claiming that he spoke with somebody who knows the person who posted these pictures and claims that it's a defective unit directly from Foxconn.
All units were made at Foxconn afaik, we know of the FCC filing being approved on August 3rd, this is likely that device, and if they can get a device approved in august, that means it can be ready for the july devkit, the SoC was finished on july 18th-25th, there is no way you'd punch out 1000+ of these devices and sit on them for 3 to 4 months for final devkits, besides there is no reasonable point to use 2 32bit memory modules to hit 64bit bandwidth, or use a separate chip for the flash memory on final hardware, this looks like a devkit honestly.
Whoa, settle down folks.
The thing that jumps out most immediately to me at an architectural level is the fact that the battery is a single discrete component rather than a conglomerate of smaller modules that are more evenly dispersed for heat and balance reasons. Tablets today have followed that design paradigm for the better part of a decade and I'm confused why that isn't the case here for something that has that form factor. I'd like to see more of it before I say any more but that's my take from a distance. I can't really speak to the hardware specs themselves until more is known about them.
I remember a discussion about how devkits are not usually manufactured in the same place as the retail ones, but I can't seem to find it right now.
Besides that, we don't know when this unit was manufactured only the SoC, so being the FCC test unit is very speculative.
What about this design doesn't look like the july devkits to you? it fits everything we know about that device. It seems silly to produce a chip in july, as you are putting together devkits to go out and not use the new chips and instead wait 3+ months. Or does that sound right to you? what logic makes yours more sound than mine about this being an october devkit that would have some boost to performance over what we know about the july devkits and why would they sit on it for 3 months?
Begs the question, why this small? They could have made a bigger tablet
The logic in which the logistics to manufacture chips, deliver them to Foxconn, manufacture the units, deliver them to NA or Europe and then deliver them to the developers takes more than 2 weeks. Plus we had reports that the july devkits have a standard X1 in them, not a custom one, no?
And seeing how this has only 32GB of storage, that makes this very unlikely to be a devkit. Maybe it's just a retail unit that broke down during the QA at Foxconn. Could be part of the trial production run that we had reports about happening in Q3 2016. There are so many possibilities, that stating something as a fact or very sure seems not very wise at this point.
To be fair I'd think that a July dev kit would be one produced in July, not necessarily reaching everyone in July.
This still baffles me: what would they gain by removing those cores?! Especially with the leaked documentation implying 1 "main" core reserved (likely for the OS, right?)space saved by removal of four Cortex A53s
So basically any USB charger could work? This might be interesting when gaming on the go, it would be nice to be able to connect a commonly available charger when/if needed.It doesn't need quick charge. USB-C uses PD or power delivery. PD can provide up to 100W of power through 5 voltage level.
5V/3A = Up to 15W - Most USB-C smart phone uses this and maxed out at this level for safety and temperature reasons. The Nintendo switch itself most likely charge at this level since it takes 3 hours to charge it's 4000mAH battery.
That's consistent in the text output though (allegedly from issuing a command/executable) so it might simply be a typo made by the programmer who wrote the aforementioned software.And there is a typo, core-croe. This is most likely a fake.
How long is typical soc manifacture to product release time?
Now this is new. Haven't heard that one before. I was wondering that too tbh. I wouldn't mind a bigger Switch myself but surely others find it a perfect trade-off between portability and power. I don't play handhelds in short train rides so for me it could've been bigger. But I'm saying that without having actually tried it and I'm sure Nintendo has tested different sizes and weights extensively before choosing this one.
Didn't we got the info that devs received devkits in July? Together with the documentation that leaked earlier this month?
Edit: this report from Eurogamer was on 26th of July, so some devs got it before that.
Just saying that it wouldn't require being delivered to all or even most devs in July to be called the July kit. I mean just the documentation being dated as July would be enough. But as you say obviously the kit did start to be sent out in July, exactly how early it started is unknown.
Didn't we got the info that devs received devkits in July? Together with the documentation that leaked earlier this month?
Edit: this report from Eurogamer was on 26th of July, so some devs got it before that.
The timeline is fine. Here is the logic with what he is suggestion. He is saying that they sent out devkits at the same time new chips were being produced and then sat on those for 3+ months.
There is no reason they wouldn't just wait a few extra days to get the new chips in these devkits. It still being July Devkits, just means it doesn't take as long as he is thinking to ship these to developers, I mean if the chips are done on the 18th and foxconn is making the devkit (or even Nintendo given this is a prototype, and foxconn leak confirmed that at least 1 switch was put together internally at Nintendo)
This isn't the first devkit so Foxconn could put the chips on the devkit in a single day, possibly 2 and ship them to Nintendo well before the end of july to be shipped out to developers. Nintendo not using these chips for developers make no sense.
No the X1 chip was custom even then, just think about it, this is Revision #2, there was an A0 and A1, so they would all be custom Maxwell GPUs based on X1.
I'm saying that this is no devkit at all anyhow. Because it has a 32 GB storage, not 64GB as the devkits.
We know that the final devkits have 64GB of storage, all rumors before that only suggested 32GB of internal storage actually.
We know that the final devkits have 64GB of storage, all rumors before that only suggested 32GB of internal storage actually.
The July devkit documentation clearly states 64GB storage.
Is there a link for that? I thought we only found out because of european commercial not properly covering the debug menu with gameplay.
It's in these documents.
http://www.neogaf.com/forum/showthread.php?t=1343614
Direct link to the relevant page: http://dystify.com/Overview/contents/Pages/Page_94755783.html
I think if they decided to go with Maxwell instead of Pascal that's a really bad decision, Pascal would improve battery life and heat, which is really important for a "portable".
I really hope it's Pascal, even if it's a custom version of X1 instead of X2.
You mean here, where is says 32GB?
Wait, I thought you were talking about the dev kits and not the retail unit?
I was, I thought that outlined the devkits. I wonder if the FCC prototype would use the 32GB storage, since it is removable, it could be an easy solution to this problem. The timing is hard to believe for a retail unit, those just wouldn't have been made yet and we know there was a change in october, that being a chip they sat on for 3 months makes 0 sense.
I was, I thought that outlined the devkits. I wonder if the FCC prototype would use the 32GB storage, since it is removable, it could be an easy solution to this problem. The timing is hard to believe for a retail unit, those just wouldn't have been made yet and we know there was a change in october, that being a chip they sat on for 3 months makes 0 sense.
It could be part of the trial production that was rumour to take place in the Q3 2016. Would fit the timing perfectly.
The prototype for the FCC makes the most sense, a trail run happened in October actually afaik, not before August 3rd. Final hardware actually wasn't ready until October either from rumors.
Also isn't Q3 usually referring to October, November and December? even "Fall Quarter" starts at the very end of September usually right?
Wish someone would just take a final retail unit apart. Personally if I was that guy who got his Switch early I'd have sold it to DF, can't even play the thing for two weeks anyway.
Everybody wants a 1000TFLOPS system with Nintendo games that have $1B budgets, but we just don't think it's feasible market or technology wise. And given the constraints and the competition, I'd rather have Sony focus on big blockbuster stuff, and Nintendo on creative play solutions, than the other way around. Within this framework though, we can want Nintendo to be as powerful as possible, and Sony to be as creative as possible. But it's not likely, or perhaps even preferable, that one will shift over into the other's expertise area.Just out of interest, why do people repeatedly say they don't want a powerhouse console from Nintendo, that they would prefer they did their own thing and then spend many, many pages trying to work out if Nintendo have some hidden hardware, or if their hardware is actually more powerful than it first seems?
They did the same for the Wii U and now the Switch, I thought hardware power didn't matter?