frankie_baby
Member
Wut
My 5 year old galaxy s3 screen still looks great.
My galaxy nexus had the keyboard permanently ghosted on it after 4 months and i my mrs' s3 did too but cant remember how long that took
Wut
My 5 year old galaxy s3 screen still looks great.
With the joycons and HD rumble being part of the system is there any viable way to make a clamshell Switch work and be stable. Honestly, I don't believe there is any reason Nintendo would do this given all the positive idea provoking responses we have seen from developers about HD rumble. I think the 4K dock has more plausibility than a smaller/clamshell Switch.Moving to a clamshell and putting the battery there would give plenty of room to make a smaller device. The tablet itself is smaller than a n3DSXL anyways.
I had an issue with my S3 where it would not go to sleep. The screen stayed on all day for at least 14 hours a day, for a week straight(I took the battery out at night). Got it fixed and had no screen burn in.My galaxy nexus had the keyboard permanently ghosted on it after 4 months
I had an issue with my S3 where it would not go to sleep. The screen stayed on all day for at least 14 hours a day, for a week straight(I took the battery out at night). Got it fixed and had no screen burn in.
Just to understand, you're saying that this is from the FCC prototype and July devkits but the retail units use a different SoC despite the FCC docs stating that the prototype it's equivalent to the mass produced items?
What are you expecting to be in the retail units?
With the joycons and HD rumble being part of the system is there any viable way to make a clamshell Switch work and be stable. Honestly, I don't believe there is any reason Nintendo would do this given all the positive idea provoking responses we have seen from developers about HD rumble. I think the 4K dock has more plausibility than a smaller/clamshell Switch.
Help me understand if you think I'm off the deep end here.
My galaxy nexus had the keyboard permanently ghosted on it after 4 months and i my mrs' s3 did too but cant remember how long that took
Are you sure it wasn't a counterfeit S3 ?My galaxy nexus had the keyboard permanently ghosted on it after 4 months and i my mrs' s3 did too but cant remember how long that took
Sony is not creative ?
Games like TLG,BB,Nioh or Horizon aren't creative ?
Is creativity limited to cutesy low budget games or tacked on gimmicks by Nintendo ?
Are you sure it wasn't a counterfeit S3 ?
My S3 still works perfectly after 4,5 years.
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.
HD Rumble works in the pro controller, why could it not work in a clamshell design? Joycons being separate only make sense in table top or tv mode which makes a pro controller style clamshell make perfect sense for a handheld.
If they're lucky, maybe they can stick a small kinetic cooler in there and get a performance boost via less overheating:I wonder what they'll change for a Switch revision. More flash memory obviously, probably bigger screen. They can't do much more right?
I don't know, man .. the system clocks, GFLOPS numbers, cores, etc seem so random at a glance. Why would they play around with clock frequency out-of-sync with the system clock? I'll give it another look later. Where did that screen come from?It's not x1, this isn't the retail switch, it's something else if real. So those clocks don't apply.
I don't know, man .. the system clocks, GFLOPS numbers, cores, etc seem so random at a glance. Why would they play around with clock frequency out-of-sync with the system clock? I'll give it another look later. Where did that screen come from?
its all crazy. Again the report we got is Capcom asked for more storage/RAM NOT PROCESSING POWER. so obviously whatever the switch got going on it can run what is needed. We still haven't heard a developer shit on the console like multiple ones did with the Wii U. something is not right.
Do we already know if the tegra in the new shield throttles like the one in the older model?
You dont know what capcom actually asked for vs what they got. They obviosly will put a positive spin when talking publicly about a system they want to sell games on. Maybe they only mentioned memory because thats the area Nintendo bulged on.
Yeah. They also did a benchmark test on many systems, and the PSVita wasn't DESTROYED as much as it should have compared to the Switch.I was on my phone and only did the first lower clocks, I was factoring in the CPUs GFLOPs. It's wrong as the second set of clocks give far too high a gflops number to match up with what was written. It's 100% fake IMO.
its all crazy. Again the report we got is Capcom asked for more storage/RAM NOT PROCESSING POWER. so obviously whatever the switch got going on it can run what is needed. We still haven't heard a developer shit on the console like multiple ones did with the Wii U. something is not right.
Speaking of that, the Capcom representative during that devkit meeting reportedly referred to the Switch as "High performance and low power," more than once, and this wasn't a PR statement. It sounds like they were very impressed with the power.
I had an issue with my S3 where it would not go to sleep. The screen stayed on all day for at least 14 hours a day, for a week straight(I took the battery out at night). Got it fixed and had no screen burn in.
A single memory chip module and soldered flash memory. If the chip isn't 16nm, I think we would get that to be able to do the foxconn tests for 8 days straight, as a 20nm x1 based SoC wouldn't be capable of such a test in the Switch, with it's cooling. Nothing that would violate the FCC filing, just small things, it's mostly timing I have a problem with, as july devkits would have been august devkits and final since software was the focus with the device, instead we have july devkits and no new devkit until october.
This. The FCC aren't going to hang them out to dry because they've (potentially) moved from 20nm/A53 to 16nm/A72. The submitted product would still continue to represent the functional, material purpose of the final retail unit for all intents and purposes.
Did many developers complain about the 3ds?
I think the developers acknowledge the limitation of the tablet form and appreciate the tech considering this.
Just a note. A72/16 nm is not a leak, but a speculation within the Foxconn leak, based on the clocks.
And it's A57 in July development docs, not A53.
The high GFLOPS, the bash script and the mere 10C difference between standby and full-power do cause a serious doubt of the truthfulness of this screengrab. Also, I haven't seen an ARM cluster yet where cores can be individually clocked, but that could be a matter of performance governor after all.I was on my phone and only did the first lower clocks, I was factoring in the CPUs GFLOPs. It's wrong as the second set of clocks give far too high a gflops number to match up with what was written. It's 100% fake IMO.
I had never seen this....Atleast there seems to be no wasted space this time.
The high GFLOPS, the bash script and the mere 10C difference between standby and full-power do cause a serious doubt of the truthfulness of this screengrab. Also, I haven't seen an ARM cluster yet where cores can be individually clocked, but that could be a matter of performance governor after all.
Capcom specificly mentioned comparisons to current gen platforms. Nobody was trying to downport games from more powerful platforms to 3ds.
What games is Capcom downporting to Switch now?
I have no idea, but ram amount compared to current gen woudnt matter if they didnt want to run current gen games on it anyway.
Yes, I'm heavily leaning towards fake.So you think it is fake too? Or is there a little chance of it being true?
The 800 Gflops number seems to be so damn high.
"High performance and low power," has been an internal Nintendo hardware slogan since the gamecube. Go read Iwata Asks for Wii or Wii U. Takeda also used the same words in the recent Q&A.
One thing thats very odd about the X1 with no A53 theory is that both Nintendo and Nvidia insists its a custom soc. If its 99% X1, why be so secretive about it?
Capcom was an exception, though. They ported down SFIV to the 3DS.Capcom specificly mentioned comparisons to current gen platforms. Nobody was trying to downport games from more powerful platforms to 3ds.
RUMOR is RE7 is being ported, but still a rumor. I would like to point out that the source did get other rumors right like online not being a fee "at launch" and making the wording very clear about it being charged later. Also some other things, it was part of that Obe1's Switchmas thing.
He actually did mention some very specific things that ended up being true about the presentation when even LKD was off on it.Phew...he also said that the Switch can be charged in 30 minutes and stuff like that. I highly doubt he knew anything from the beginning.
He actually did mention some very specific things that ended up being true about the presentation when even LKD was off on it.
One thing thats very odd about the X1 with no A53 theory is that both Nintendo and Nvidia insists its a custom soc. If its 99% X1, why be so secretive about it?
If it's really is a 64bit bus, that's certainly custom and would be great.Nintendo doesn't really want the specs to be released, and since it's their chip Nvidia has to remain silent as well.
About the A53 fact is that it seems there were already X1 without them already in some products, I did some digging a while ago and some wiki Tegra pages reflected that, I'll try to link that later since I'm on mobile now.
So recapping some info.-
* July docs: 4A57s (3 usable for games)
* No A53s
* Chinese photos pointing to a 64bit bus
* 1600 Lpddr4
The question again is, what is customized if even no A53s isn't really new?
About the clocks 4SM thing, as pointed out there are too many things contradicting that possibility, plus by my estimations SoC being in a similar fab mode would be around 150ish mm which doesn't seem to be the case looking at the chinese photos.
We still need some die shots from both Switch and Tegra X1 (the one on the slides is just a PR render) and some hacking to know the clocks, but things seem to be moving really fast and launch is almost here so hopefully we won't have to wait too much.
Really? What were these specific things? Don't remember anything tbh.^^
Regarding where this unit came from, the FCC test unit makes no sense to me. Why would the FCC, a United States governmental organization, give the test unit to someone who then somehow gets it into the hands of a Chinese company that does electronics tear downs? Does the FCC give it back to Nintendo? Do they store their test units? I can't imagine a third option here, and if it's either of the first two I don't really know how this source from the OP could have gotten the unit.
It makes far more sense as a defective unit from Foxconn, devkit or not. Likely not devkit due to the 32GB of storage. These images really don't say much about the SoC regardless, so the Foxconn clocks/16nm could still be accurate. The only thing we can reasonably determine from these figures is the amount of RAM and bandwidth, and clearer pictures would probably be more definitive there.
And did that readout about the clocks and flops with 516 cores come from this same source? If so that's confusing, as that looks quite fake.
If it's really is a 64bit bus, that's certainly custom and would be great.
And did that readout about the clocks and flops with 516 cores come from this same source? If so that's confusing, as that looks quite fake.
It's not the fcc unit, it's a prototype, like the one they made for the fcc. Nintendo made the switch internally according to the Foxconn leak, we don't know how many, but the time lines don't match up. Making an unfinished devkit and within a week, make a functional final device, give it to the fcc and then wait 3 months before giving developers final devkits? How does that work?
That fake picture is not from the same source as the Switch SoC.
Turn your screen sideways, turn the brightness up and tell me you don't have a band of slightly different hue at the side of your screen, where the Android taskbar goes (unless you're color blind). I've had multiple Galaxy phones and they ALL had this. It's not burn in by the way, but depletion, and it's simply a factor of OLED. A slight price to pay for the superiority of OLED in terms of contrast and speed outside of this, but I'm fine with gaming consoles not having it.
I don't know how long the gap typically is between SoC being fabbed and the final device being sent out, whether as a devkit or retail unit so I can't answer that. It could be that it's usually 3-4 months, which could mean the July devkit SoCs were fabbed in April or March and the October SoCs (which could also be retail SoCs) were fabbed in June-July. Without knowing how this is typically done (both by Nintendo and others in the industry) we can't really make judgements about the timeline.
You could be right that it's an old prototype, but I don't think the timeline is any reason to disqualify it from being a final unit.
As I pointed out before, the new Shield from Nvidia has a SoC manufactured 3 weeks later than Switch's SoC and that was just released. So several months between SoC manufacturing and release doesn't really sound far fetched.