If you disagree, then please explain what about this device makes it a handheld rather than a console. It's a capable gaming device much more powerful than the previous generation device, designed to be played hooked up to a TV using controllers not connected to the device, that is going to be playing games designed for consoles.
Itsn't happening, let it go
I think this could be one of the 100 other options. Dont bring your opinion like a fact.Can we stop responding to Malo when he's acting like a bitch? He's making these threads impossible to enjoy and this just makes it worse.
Emily seemed to imply later that it wasn't really close to Xbone, though. Either way, better than Wii U in a handheld still sounds good. They worked miracles with the Wii hardware, but I think most people here would be lying if they said they weren't wishing for a bit more. Prayin' for Pascal.
Iirc, your worst case scenario was a 64 gflop handheld, so I don't think we're quite there yet.
Well then, I'm glad I didn't go with my original yoga pants joke.
Saying what makes it a good handheld isn't saying what makes it NOT a console. As a hybrid device it is both. If you can tell me what features it is missing for being a console, then you would have a point. As for your "every phone is a hybrid", what game controls do they have built-in? Are they almost as powerful as a PS4/XBox One? Are they designed for use on a TV, or is that capability thrown in as an extra feature? Are they getting modern brand new console games?
Emily Rogers said:
Who cares? The end result is that it's a pos as console thus only fit to be a handheld.
The detachable controllers for local multiplayer, they are enough to consider it a hybrid somehow. Not a fan of the whole concept, in fact this detachable controllers sound like a less than elegant solution.
Well, I did, but you may not have seen it since it was in an edit: As for your "every phone is a hybrid", what game controls do they have? Are they almost as powerful as a PS4/XBox One? Are they designed for use on a TV, or is that capability thrown in as an extra feature? Are they getting modern brand new console games?
Well, "at time of launch" is incorrect, the PSP didn't get TV-out abilities until the 2000 model, and even that one didn't ship with the AV cable it had to be bought separately. Also it was designed to only play handheld games, the only console games it got were ports/emulations from previous generations, so it didn't fit the "modern console games" rule (although Nintendo really is changing things up in this regard, since they don't have a difference between handheld games and console games anymore, they reorganized to integrate their handheld developers with their console developers).
If it's designed/marketed as a primary use of the device. I have yet to see a phone marketed with TV out as a primary feature - these days it's just a bullet point, if even that. Not to mention they don't even ship with the HDMI cable needed to hook up to a TV.
Yup. Also osirisblack's comment about ports having little to no issue running on NX.
Read the post he was responding to.
No such thing was said.
We're never going to see eye-to-eye on this, so think of it this way: Not everyone cares about a powerful handheld, and not everything thinks another GC->Wii (though a bit better) kind of jump is acceptable.It's not a diffibult concept. Some people want more than you do, and just telling them to be positive because you like the concept isn't going to make them change what they want or expect. If it's a console, then it's a shitty console with a shitty default controller and a power level likely far too low for its price. Nintendo doesn't exist in a vacuum, and as such should not be judged as i9 Wii U is the only console on the market. Further, even as a Wii U successor this is a poor jump, possibly smaller than the jump from PS4 to Neo.
That would be half as powerful as Wii U. How weak did you think Wii U was?
The bolded isn't true. In fact, pretty much the opposite is true, as in a tightly thermally constrained environment (i.e. a handheld) the marginal benefit to increased parallelism (i.e. more SMs) can be quite large.
To demonstrate, let's look at a power curve for Pascal I've put together. Unlike my previous power curves for A72 and A53 CPU clusters (which are based on solid real-world data from Anandtech and should be considered reasonably accurate), this is a much more rough approximation based on just four data points:
- TSMC's claims of "40% higher speed" and "60% power saving" over 20nm, each applied separately to the TX1's GPU drawing 1.5W at 500MHz (divided by 2 for 750mW per SM).
- Power draw readings from the GTX1080 before and after overclocking (full board power readings, minus GDDR5X, divided by number of SMs).
Obviously I'm extrapolating a lot from fairly poor data, but hopefully it should be in the right ballpark, and enough for our discussion in any case. (I should also note that this isn't strictly a measure of power draw for the SMs themselves, but rather a measure of the draw of an entire Pascal GPU "per SM", so including other components like ROPs, TMUs, etc., assuming they're always in roughly the same proportion to SMs). In any case, here's the power curve:
The important thing to note is that, like virtually all IC power curves, it's not linear, and for a given increase in clock speed you require a much larger increase in power consumption to get you there. What this means is that you'll get better performance by using more SMs at a lower clock speed than fewer SMs at a higher clock speed.
Let's look at the clock speed (and raw floating point performance) that could be achieved with different numbers of SMs within the power constrains we might expect for a handheld GPU:
1x SM:
1000 mW - 780 MHz - 200 Gflops FP32 - 400 Gflops FP16
1500 mW - 915 MHz - 234 Gflops FP32 - 468 Gflops FP16
2000 mW - 1025 MHz - 262 Gflops FP32 - 525 Gflops FP16
2x SM:
1000 mW - 595 MHz - 305 Gflops FP32 - 609 Gflops FP16
1500 mW - 700 MHz - 358 Gflops FP32 - 717 Gflops FP16
2000 mW - 780 MHz - 400 Gflops FP32 - 800 Gflops FP16
3x SM:
1000 mW - 510 MHz - 392 Gflops FP32 - 783 Gflops FP16
1500 mW - 600 MHz - 461 Gflops FP32 - 922 Gflops FP16
2000 mW - 670 MHz - 515 Gflops FP32 - 1030 Gflops FP16
As you can see, a 3x SM configuration can achieve nearly the same performance with 1000mW that a 2x SM configuration can with twice that, and a full 50% more than a 1x SM config can manage with 2000mW at hand.
This isn't to say that I expect a 3x SM GPU in the NX, but there would certainly be a sizeable performance jump over 2x SMs if they decided to do so.
The bolded isn't true. In fact, pretty much the opposite is true, as in a tightly thermally constrained environment (i.e. a handheld) the marginal benefit to increased parallelism (i.e. more SMs) can be quite large.
To demonstrate, let's look at a power curve for Pascal I've put together. Unlike my previous power curves for A72 and A53 CPU clusters (which are based on solid real-world data from Anandtech and should be considered reasonably accurate), this is a much more rough approximation based on just four data points:
- TSMC's claims of "40% higher speed" and "60% power saving" over 20nm, each applied separately to the TX1's GPU drawing 1.5W at 500MHz (divided by 2 for 750mW per SM).
- Power draw readings from the GTX1080 before and after overclocking (full board power readings, minus GDDR5X, divided by number of SMs).
Obviously I'm extrapolating a lot from fairly poor data, but hopefully it should be in the right ballpark, and enough for our discussion in any case. (I should also note that this isn't strictly a measure of power draw for the SMs themselves, but rather a measure of the draw of an entire Pascal GPU "per SM", so including other components like ROPs, TMUs, etc., assuming they're always in roughly the same proportion to SMs). In any case, here's the power curve:
The important thing to note is that, like virtually all IC power curves, it's not linear, and for a given increase in clock speed you require a much larger increase in power consumption to get you there. What this means is that you'll get better performance by using more SMs at a lower clock speed than fewer SMs at a higher clock speed.
Let's look at the clock speed (and raw floating point performance) that could be achieved with different numbers of SMs within the power constrains we might expect for a handheld GPU:
1x SM:
1000 mW - 780 MHz - 200 Gflops FP32 - 400 Gflops FP16
1500 mW - 915 MHz - 234 Gflops FP32 - 468 Gflops FP16
2000 mW - 1025 MHz - 262 Gflops FP32 - 525 Gflops FP16
2x SM:
1000 mW - 595 MHz - 305 Gflops FP32 - 609 Gflops FP16
1500 mW - 700 MHz - 358 Gflops FP32 - 717 Gflops FP16
2000 mW - 780 MHz - 400 Gflops FP32 - 800 Gflops FP16
3x SM:
1000 mW - 510 MHz - 392 Gflops FP32 - 783 Gflops FP16
1500 mW - 600 MHz - 461 Gflops FP32 - 922 Gflops FP16
2000 mW - 670 MHz - 515 Gflops FP32 - 1030 Gflops FP16
As you can see, a 3x SM configuration can achieve nearly the same performance with 1000mW that a 2x SM configuration can with twice that, and a full 50% more than a 1x SM config can manage with 2000mW at hand.
This isn't to say that I expect a 3x SM GPU in the NX, but there would certainly be a sizeable performance jump over 2x SMs if they decided to do so.
