New PS4 Smell

[Shifty Geezer]

I thought it was odd at first too. I'm quite sure there's a technical explanation for it, I'm just not in any position to provide that explanation.

Perhaps standby boot multitasks operations and thrashes the drive head, whereas cold boot loads data linearly and more optimally for the HDD? Clearly active downloads in the queue preserves more in RAM. I'm guessing presently the standby is flushing everything, or not registering it as already loaded, and can be fixed in FW.

 

[Gradthrawn]

Gradthrawn does fan spin while the console in the awake standby mode (when there are some downloads in progress)?

Nope.

Actually, let me clarify that. At my seating position, about 4 feet away, I cannot hear anything from the console once it goes into standby mode with active tasks running in the background. However, I have not held my head to the console under those conditions to verify if the fan is running or not. That said, please take my perceived noise level experiences with a grain of salt. I've heard people refer to the PS4 as "loud". With a full game running, a browser open, and a media app open in the background I can hear an audible whir from the fan. Anything less than that and its quiet for me even during silent moments. This is with a 72 degree F ambient room temperature.

 

[scf]

Thanks Gradthrawn

 

[Gradthrawn]

Perhaps standby boot multitasks operations and thrashes the drive head, whereas cold boot loads data linearly and more optimally for the HDD? Clearly active downloads in the queue preserves more in RAM. I'm guessing presently the standby is flushing everything, or not registering it as already loaded, and can be fixed in FW.

Hmm, that's definitely possible. However, I'm note sure that lines up with the relative performance differences I'm seeing from stock to SSD for each boot mode. The stock Z5K500 should be able to hit around 100 MBs sequential read. The Samsung 830 I have installed, even if artificially capped at 150 MBs as has been alluded to, should still hit 150 MBs in sequential everything even on a SATA II bus.(granted that test I linked to was with the newer 840 Pro on a SATA III controller manual configured to SATA II speeds). In other words, I would think an additional 50 MBp/s of sequential read speed would garner more than 1 second shaved from a cold boot, if the cold boot is optimized for sequential read as you're guessing. Of course, we have no real way of testing the internal transfer speeds at this point. It could very well be capped to 125 or even ~100 MBs. Further, there could be other limiting factors.

 

[Silent_Buddha]

With a cleared download queue, and all games fully updated (ie no pending background tasks) I rested the start-up with the replaced drive (drive specs noted above:

From a Cold Boot: 21 seconds (video to display at about 16 - 18 seconds, at login screen at 21 seconds)
From stand-by: 23 seconds.(video output to display at about 16 seconds, at login screen at 23 seconds)

So the new drive shaved about 5 seconds from standby but only about 1 (1.5 at best) from a cold boot. Each test run 3 separate times. Times recorded from the boot "beep". Previous results below:

HGST Z5K500-500 (Original Drive):
Cold Boot: 22 seconds
Standby, 29 seconds.

Samsung 830 (256 GB):
Cold Boot: 21 seconds
Standby: 23 seconds

Samsung 830 (256 GB): (with active downloads in queue)
Cold Boot: 21 seconds
Standby: 3 seconds

Looking at those numbers, it would seem that what Sony are doing is what would be called Hibernate (memory saved to disk, virtually everything unpowered) on a Windows machine unless a download is active at which point it is more similar to Sleep mode, except with a few differences. So in addition to memory being powered as in sleep mode, the ARM chip is probably also kept powered on while the rest of the system is unpowered.

That would explain why Standby without active download takes longer than a cold boot. A thin OS like a console OS isn't going to impact cold boot times in the way a full OS would (although oddly, minus BIOS on power up that's slower to boot than Windows 8). But if you're having to also load up to 8 GB of data off the HDD at the same time that's going to make boot take a tad bit longer.

Regards,
SB

 

[patsu]

They said they are still working on suspend/resume. So this is perhaps a snapshot of that progress.

 

[scf]

As I understood, they meant that you can't suspend a running game.

 

[Gradthrawn]

Looking at those numbers, it would seem that what Sony are doing is what would be called Hibernate (memory saved to disk, virtually everything unpowered) on a Windows machine unless a download is active at which point it is more similar to Sleep mode, except with a few differences. So in addition to memory being powered as in sleep mode, the ARM chip is probably also kept powered on while the rest of the system is unpowered.

That would explain why Standby without active download takes longer than a cold boot. A thin OS like a console OS isn't going to impact cold boot times in the way a full OS would (although oddly, minus BIOS on power up that's slower to boot than Windows 8). But if you're having to also load up to 8 GB of data off the HDD at the same time that's going to make boot take a tad bit longer.

Regards,
SB

Ah, that would make sense.

Just noticed there's no Wireless or Wired Content Management support between PS4 and Vita. Connect a Vita via USB and the PS4 reports that "this device is not supported." Does not show up when trying to add a Wireless Device from the Vita, either. Although that last one may be due to my local network. Remote Play works fine, however.

 

[Shifty Geezer]

Is there a different switch off time for power off and sleep to account for 8 GBs being saved to disk?

 

[Gradthrawn]

Is there a different switch off time for power off and sleep to account for 8 GBs being saved to disk?

Good question. Both are quite slow. I'll test and report the results later.

 

[Silent_Buddha]

Is there a different switch off time for power off and sleep to account for 8 GBs being saved to disk?

If Sony are playing it safe and using a "hybrid sleep" mode when active downloads are in progress, then memory may be saved to disk for that as well. For example, in case someone unplugs the PS4.

Regards,
SB

 

[deathindustrial]

The web browser has a "delete history" option which appears to be completely independent of the "frequently used pages" list. Those you can only delete one by one for some unknown reason.

Seems like a potential vector for semi-permanent exploits.

The browser is OK though I have not yet found a site with HTML5 video to test.

Cheers

 

[patsu]

As I understood, they meant that you can't suspend a running game.

Yeah, how does that work ?

Will it be used if I sent my PS4 to standby ?

 

[ERP]

As I understood, they meant that you can't suspend a running game.

FWIW that was always a planned feature, but a lot of none essential items were pushed out past launch.

 

[patsu]

Ah, that would make sense.

Just noticed there's no Wireless or Wired Content Management support between PS4 and Vita. Connect a Vita via USB and the PS4 reports that "this device is not supported." Does not show up when trying to add a Wireless Device from the Vita, either. Although that last one may be due to my local network. Remote Play works fine, however.

So no more local syncing between PS4 and Vita ? Everything route through Sony cloud ?

 

[temesgen]

You can't base network capabilities off the first week of PS4 release.

One of the biggest critism of PSN for VITA and PS3 has been download times and IMO it sends the wrong message for PSN on PS4 which is now largely a paid service to have the same sort of issues.

Sony knew they had a launch day OS update and they also know that there a number of apps and games people will try to download. It will literally be days before people are able to download some of this stuff.

And it gets even worse bc many of the games using MP are struggling to get connections so while I agree with you that the traffic demands are perhaps higher than normal its also true that Sony should be expecting more traffic moving forward and they should have the infrastructure in place to make sure people have a good experience and want to use the cloud more. Sadly all they have done here is give gamers an uneasy feeling about what they might be able to expect.

 

[deathindustrial]

Controller battery life is a lot worse than the PS3, possibly due to headset use. I haven't timed it but it seems to be around 3 hours.

 

[patsu]

Controller battery life is a lot worse than the PS3, possibly due to headset use. I haven't timed it but it seems to be around 3 hours.

There is a GAF thread claiming that it's 8 hours, vs 30 for DS3.

I haven't verified this myself.

Edit: Also... Someone confirmed that voice command over the built-in headset works.

 

[betan]

Controller battery life is a lot worse than the PS3, possibly due to headset use. I haven't timed it but it seems to be around 3 hours.

Lightbar might be a big issue.

 

[eastmen]

if only you could quickly change the batteries . 3 hours is way ot little , 8 hours isn't so bad though