Benetanegia
Regular
I'd bet on "Nintendo wasn't willing to pay for it". I think that the fact that the 2017 Shield Android TV uses TX1 is a consequence of the Nintendo deal, because I don't see why Nvidia wouldn't use TX2 in the new device.
- Status
I'd bet on "Nintendo wasn't willing to pay for it". I think that the fact that the 2017 Shield Android TV uses TX1 is a consequence of the Nintendo deal, because I don't see why Nvidia wouldn't use TX2 in the new device.
Parker would make for a natural progression for Nintendo to release Light Switch* later down the road for folks to upgrade, and by then, 16nmFF will have come down further in price anyway. Nintendo wouldn't have to pay much for ASIC R&D, and perhaps would only just pay for licensing the chip design & fabrication.
*Charger & dock not included.
Mobile mode would then feature lower power consumption for OG Switch games, then options for devs to utilize higher clocks for new games.
Docked mode is easy enough since they already have a framework for GPU clocks. Whether they expose increased CPU clocks down the road ala PS4 Pro boost for PS4 games remains to be seen, I suppose.
Denver being disabled in Max-Q is a funny coincidence, n'est pas. Memory contention with Denver + A57 + GPU would be interesting to see, but Denver may simply be off-the-table just like how the A53s are disabled for Switch, and that Max-Q mode is essentially tailored for the next Switch revision.
*Charger & dock not included.
Mobile mode would then feature lower power consumption for OG Switch games, then options for devs to utilize higher clocks for new games.
Docked mode is easy enough since they already have a framework for GPU clocks. Whether they expose increased CPU clocks down the road ala PS4 Pro boost for PS4 games remains to be seen, I suppose.
Denver being disabled in Max-Q is a funny coincidence, n'est pas. Memory contention with Denver + A57 + GPU would be interesting to see, but Denver may simply be off-the-table just like how the A53s are disabled for Switch, and that Max-Q mode is essentially tailored for the next Switch revision.
Last edited:
D
Deleted member 13524
Guest
Parker would make for a natural progression for Nintendo to release Light Switch* later down the road for folks to upgrade, and by then, 16nmFF will have come down further in price anyway. Nintendo wouldn't have to pay much for ASIC R&D, and perhaps only just paying for licensing the chip design & fabrication.
*Charger & dock not included.
Yeah, when the rest of the world has moved past 10nm and every SoC out there is using 7nm with post-A73 cores give out over 4x better power efficiency.
There's a chance Tegra X2 is significantly larger than Tegra X1, though.
In terms of density, 20nm is close to 16FF. ISP and video blocks aside, we see the TX2 dropping a small module of Cortex A53 while gaining a pair of Denver cores which are pretty big, plus doubling the system memory interface.
Parker would make for a natural progression for Nintendo to release Light Switch* later down the road for folks to upgrade, and by then, 16nmFF will have come down further in price anyway. Nintendo wouldn't have to pay much for ASIC R&D, and perhaps would only just pay for licensing the chip design & fabrication.
*Charger & dock not included.
Mobile mode would then feature lower power consumption for OG Switch games, then options for devs to utilize higher clocks for new games.
Docked mode is easy enough since they already have a framework for GPU clocks. Whether they expose increased CPU clocks down the road ala PS4 Pro boost for PS4 games remains to be seen, I suppose.
Denver being disabled in Max-Q is a funny coincidence, n'est pas. Memory contention with Denver + A57 + GPU would be interesting to see, but Denver may simply be off-the-table just like how the A53s are disabled for Switch, and that Max-Q mode is essentially tailored for the next Switch revision.
And Nintendo has conditioned their users for this already. They had the ds then ds lite , 3ds then new 3ds and that's not including the xl editions .
IF MS and Sony stay out of this market I would spend $300 on a switch 2 with an updated soc in 2019/20 and I am sure a lot of others would too. Esp if Nintendo keeps compatability with games and accessories
D
Deleted member 13524
Guest
In 2019/2020 you'll have x86 7nm SoCs consuming ~5W-10W capable of using 128bit LPDDR4X (or even LPDDR5?) and/or low-cost HBM.
All it takes is a Smach Z follow-up to put one of those new SoCs into a handheld format and you'll get your $300 much better spent elsewhere.
In fact, I do think even the current Smach Z at $330 is a much better value proposition than the Switch is at $300. I have tens of somewhat older games that would run like a charm in it.
Assuming it actually comes out, of course...
CaffeinatedChris
Newcomer
I'd like to see the Chinese tablet makers (Onda, Chuwi, Teclast, et al) take a stab at a Switch form factor tablet PC.
They can already give us an 8" tablet with a 1080p IPS display, x5-8300, 2GB RAM, 32GB eMMC, and decent battery life for USD$100.
Double the RAM and NAND, bump us up to the x7, shrink the screen, and make it thicker to accommodate the battery. Include a couple clip-on BT controllers, charge USD$200 - hell, maybe more - and call it the Change instead of the Switch.
The GPD WIN is interesting, but I'm not a fan of the clamshell form factor (and what's with the control stick positioning?)
A 12EU Cherry Trail GPU @ 600MHz would put out 12*16*500 = 112.5GFLOPs. 16EU in the x7 would get to 150GFLOPs, or roughly the same as an undocked Switch.
And honestly, look at some of the Switch's most-hyped indie titles - Shovel Knight, Stardew Valley, Steamworld Dig 2 - it's not like they couldn't run on that much power.
They can already give us an 8" tablet with a 1080p IPS display, x5-8300, 2GB RAM, 32GB eMMC, and decent battery life for USD$100.
Double the RAM and NAND, bump us up to the x7, shrink the screen, and make it thicker to accommodate the battery. Include a couple clip-on BT controllers, charge USD$200 - hell, maybe more - and call it the Change instead of the Switch.
The GPD WIN is interesting, but I'm not a fan of the clamshell form factor (and what's with the control stick positioning?)
A 12EU Cherry Trail GPU @ 600MHz would put out 12*16*500 = 112.5GFLOPs. 16EU in the x7 would get to 150GFLOPs, or roughly the same as an undocked Switch.
And honestly, look at some of the Switch's most-hyped indie titles - Shovel Knight, Stardew Valley, Steamworld Dig 2 - it's not like they couldn't run on that much power.
Last edited:
Look at the wii and wiiu. now that was really old hardware, so old it was unbelievable, and it was sold at a ridiculous price, nintendo cares way more about gimmicks then they do specs, and nintendo gave up on competing with sony/MS and caring about thirdparty since the wii came out, now they have dropped out of the console race completely, and are focusing on the handheld market. they probably got an amazing deal with on the x1 that's only reason it's in the swtich, expecting the latest tech in a nintedo a console was always unrealistic based on how nintendo operates, and the rumors in july that sources from eurogamer were all pointing to a X1 in final hardware.
D
Deleted member 13524
Guest
Cherry Trail in the Atom X5/X7 series is 2 years old. Intel has since released the Apollo Lake follow-ups in the form of Pentium and Celeron SoCs. 6th gen 18 EUs at 750MHz in a 4W SDP / 6W TDP package for ~200 GFLOPs, but it supports 2400MHz LPDDR4 in a 128bit configuration, so ~40GB/s total.
Biggest problem is these SoCs are probably expensive. There are tons of Cherry Trail chips in chinese devices because Intel sells Atoms for cheap. The Atom brand is gone and Intel doesn't "subsidize" Celerons and Pentiums.
I don't disagree in terms of power. I would wager a low power zen lite as they say at 4 cores with something based on navi would be a great small portable system and could come in under $300. I've even said in other posts that MS or Sony doing this with perhaps puma+ or an evolution of puma could replicate the xbox one/ ps4 in handheld form perhaps even this year if they wanted too.
But lets face it , unless MS or Sony does it none of the companies will match the quality you get out of Nintendo and the switch. Also without a big game company behind it you wont get developers tweaking the games for the platform instead of a bunch of users posting hacks and mods and config settings to get games running a best as possible. I like that I just slide the controllers on the switch and take it with me and everything just works
The other factor is that Nintendo designs game hardware that's basically finalized a very long time before release. In order to have a good amount of time to test, develop early titles for, get end device certification, etc.
People buy Nintendo hardware to play Nintendo games, by and large, and to some varying extents to play games with their unique hardware gimmicks. They've weathered all sorts of more powerful competitors in the handheld arena, and while the entire Steam library is nothing to sneeze at there are a lot of people who aren't going to be content with that in a handheld.
tongue_of_colicab
Veteran
Not just the software IMO. Nintendo builds quality hardware too. In my experience if it's not DOA or broken soon after purchase it will probably still work 20 years from now.
That cheap Chinese stuff is fine, but just look at that Smach Z. A case that looks like its 3d printed or something and three times as thick as a Switch. Looks like it belongs in the 90's. Not saying its a bad product but its not something a company that's been doing hardware for decades would release.
That cheap Chinese stuff is fine, but just look at that Smach Z. A case that looks like its 3d printed or something and three times as thick as a Switch. Looks like it belongs in the 90's. Not saying its a bad product but its not something a company that's been doing hardware for decades would release.
I do wonder if future iterations of the Switch will use X2 or better, and what that would mean to the system. I wonder if Nintendo would be willing to do trade ins for discounts on the tablet portion only for people who'd want to upgrade. I'd also wonder if there would be games available for the Switch v2 that aren't available for v1.
It does suck that it can't get most of the big third party games. I can see stuff like Madden, Call of Duty, and a few other key games making it. But I don't know if more demanding games would be possible. How far can a game be scaled down these days? I suspect the situation wouldn't be as dire as PS360 to Wii ports, lol.
It does suck that it can't get most of the big third party games. I can see stuff like Madden, Call of Duty, and a few other key games making it. But I don't know if more demanding games would be possible. How far can a game be scaled down these days? I suspect the situation wouldn't be as dire as PS360 to Wii ports, lol.
Last edited:
Down port would be similar to XB1/PS4 -> XB360/PS3. However, since Tegra X1 has a modern GPU (with compute shaders, proper HDR formats, etc) and a modern CPU (OoO execution, cache prefetchers, etc), there would be less additional technical work. Scaling all content down would still be a huge task as the performance difference is similar to last gen vs current gen. It all depends on how successful Switch becomes and how much these cross platform ports sell. Nintendo console owners tend to prefer Nintendo games over third party games. It will be interesting to see how well the first wave of third party games sell on Switch. For example Snake Pass seems to be perfect fit for Switch customer base.
Depends on how you define significally. When you take a look at the nVidia dev kits page, booth units (X1 and X2) are pictured. The X2 seems to be a little larger than the X1. Maybe 15-25% larger per side?
I think Nintendo did not wait for the X2 because they did not want to delay the Switch further (X2 general availability, availability of sufficient X2 chips and possible last minute bugs) and they did not want to pay the higher price of the X2.
But I really see potential in the future. I could not only picture a Switch+, but also a Switch tablet with a much larger 10" screen (because I'm old and Switch screen size is no good for old people when playing Zelda in bed
).
D
Deleted member 13524
Guest
No, I don't buy this.
If the final target was e.g. a Tegra X2 using only the Cortex A57 cores, there would be nothing stopping Nintendo from sending devkits with Tegra X1 and the only practical difference would be memory bandwidth and memory amount. This is hardly a showstopper for developing 1st-gen games, especially if the final hardware is coming with better specs.
That amount of people who buy Nintendo hardware for those reasons has been counted.
It's 13.5 million sales, and it was considered an enormous failure.
If Nintendo had lifetime projections of 13.5 million sales for the Switch, they wouldn't have launched the console at all.
As soon as the reality of its performance capabilities settles in, Nintendo will be selling this as a handheld first and foremost. The post-3DS crowd will be their target, not Nintendo enthusiasts.
Of course, charging $300 for a 3DS replacement will also wear down the sales eventually.
Tegra X2 has been finalized at least since late 2015 and it has been in production at least since mid-2016.
Full assembly for the Switch's production model reportedly started in November, but the Tegra X2 had been going into Tesla cars one month earlier. And you can bet automobile QC is a lot slower and much more demanding than it is for handheld consoles.
It would have nice if nintendo made a home console similar to ps4 and sold it for 200$-249$ instead of switch. they just don't believe that third-party games can help them, cause they are established base on sony and Microsoft consoles.
as for X2 one has to wonder why no mobile device is using it.
D
Deleted member 13524
Guest
It's not hard to figure out why there's no Tegra X2 in mobile devices.
The SoC doesn't fit smartphones and Android tablets are a dying breed, with only Samsung still making new models using their own SoCs. Windows tablets need x86 CPUs and apple tablets use their own SoCs.
It's the same story for the Tegra X1 in 2015/2016. Only Google picked it up for the Pixel C.
The SoC doesn't fit smartphones and Android tablets are a dying breed, with only Samsung still making new models using their own SoCs. Windows tablets need x86 CPUs and apple tablets use their own SoCs.
It's the same story for the Tegra X1 in 2015/2016. Only Google picked it up for the Pixel C.
tongue_of_colicab
Veteran
Oh really?
The bottom one is just a render, different from the one on the linked page.
The wuublet isn't that bad either. Not the most eye pleasing design ever but the couple of times I used it it felt solid and had good ergonomics.
The original DS wasn't good looking either but somehow it is pretty cool and definitely rivalling Nokia 3310's in sturdyness. Many a times I've seen a DS phat being hurled through the room after somebody couldn't win with Mario kart hehe.
Ahhh the days were playing Mario kart with the family resulted in nearly getting murdered
Good times.
- Status
Similar threads
