Yes, H for Hybrid, your link is to the non hybrid version;
http://hdd.userbenchmark.com/Compar...-25--1TB-vs-Toshiba-MQ01ABD100-1TB/1957vsm784
http://hdd.userbenchmark.com/Compar...-25--1TB-vs-Toshiba-MQ01ABD100-1TB/1957vsm784
- Status
It's all in the H then, pretty confusing.
Toshiba also detail it
https://toshiba.semicon-storage.com/us/product/storage-products/client-sshd/mq01abxxxxh.html
I think that document is wrong. Following goonergaz's confusion, Toshiba also provide this
https://toshiba.semicon-storage.com/us/product/storage-products/client-sshd/mq01abxxxxh.html
Ok, I completed my testing;
I also did some more SSHD tests (on Pro), GTAV, U4, LoU and RotTR all (after repeat loads) got very similar to SSD speeds. I then went back to GTAV and then U4 and both stuck to the fastest time. I switched the machine off fully and then rebooted into GTAV and the time was still on par with SSD times.
The only down side to SSHD is that it takes a few loads to get these times so if playing a game which loads areas as you progress you won't get the initial fast load but certainly it's very impressive bang for buck.
My final test will be a 7200RPM on the PS4, I happen to have a spare WD Scorpio and as the PS4 is still hanging around it seems a no-brainer...will take a few days tho so don't hold your breath! lol.
I also did some more SSHD tests (on Pro), GTAV, U4, LoU and RotTR all (after repeat loads) got very similar to SSD speeds. I then went back to GTAV and then U4 and both stuck to the fastest time. I switched the machine off fully and then rebooted into GTAV and the time was still on par with SSD times.
The only down side to SSHD is that it takes a few loads to get these times so if playing a game which loads areas as you progress you won't get the initial fast load but certainly it's very impressive bang for buck.
My final test will be a 7200RPM on the PS4, I happen to have a spare WD Scorpio and as the PS4 is still hanging around it seems a no-brainer...will take a few days tho so don't hold your breath! lol.
Thx for your tests and this impressive table.Ok, I completed my testing;
I also did some more SSHD tests (on Pro), GTAV, U4, LoU and RotTR all (after repeat loads) got very similar to SSD speeds. I then went back to GTAV and then U4 and both stuck to the fastest time. I switched the machine off fully and then rebooted into GTAV and the time was still on par with SSD times.
The only down side to SSHD is that it takes a few loads to get these times so if playing a game which loads areas as you progress you won't get the initial fast load but certainly it's very impressive bang for buck.
My final test will be a 7200RPM on the PS4, I happen to have a spare WD Scorpio and as the PS4 is still hanging around it seems a no-brainer...will take a few days tho so don't hold your breath! lol.
correct me if I'm reading this the wrong way, but than there is no real difference (stock HDD) in load-times between PS4 and PS4 Pro. They are both faster in some tests and slower in others.
How does thoes differences relate to games with pro patches?
Are patched games doing more loading or other different activities, do games loading times change post patch (not sure if you can easily test this).
As I said, these are not scientific at all - and TBH I'm not 100% sure if things like background downloading might have been affecting some of the weird time differences.
My procedure; backup Pro to external HDD and use that to restore and then test all games on all the different configurations - the only exceptions being PS4 with SSHD and Pro with stock (because that's what I initially had installed before starting). The issue I had was Dead Rising was not a part of that backup so I was downloading, sometimes during testing - my bad.
What I will also add is that Uncharted 4 launch to menu on the Pro SSD is now only 16 seconds...less than half the stock times, interestingly repeat loads didn't bring GTAV or U4 to the continue times down (well, maybe a second or so, but nothing note worthy) like the launch to menu. The time is still 16 seconds after loading GTAV and then trying again, so it looks like it's here to stay.
http://www.gamasutra.com/view/news/283611/Inside_the_PlayStation_4_Pro_with_Mark_Cerny.php#tophead
I think the primitive shaders of Vega is inside the PS4 Pro GPU
http://www.anandtech.com/show/11002/the-amd-vega-gpu-architecture-teaser/2
From what they said in techreport it was designed listening complaints of console dev using GCN at low level
I think the primitive shaders of Vega is inside the PS4 Pro GPU
http://www.anandtech.com/show/11002/the-amd-vega-gpu-architecture-teaser/2
This in turn is a piece of the bigger picture when looking at the next improvement in Vega, which is AMD’s geometry pipeline. Overall AMD is promising a better than 2x improvement in peak geometry throughput per clock. Broadly speaking, AMD’s geometry performance in recent generations hasn’t been poor (it’s one of the areas where Polaris even further improved), but it has also hurt them at times. So this is potentially important for removing a bottleneck to squeezing more out of GCN.
And while AMD's presentation and comments itself don't go into detail on how they achieved this increase in throughput, buried in the footnote for AMD's slide deck is this nugget: "Vega is designed to handle up to 11 polygons per clock with 4 geometry engines." So this clearly reinforces the idea that the overall geometry performance improvement in Vega comes from improving the throughput of the individual geometry engines, as opposed to simply adding more as the scalability improvements presumably allow. This is one area where Vega’s teaser paints a tantalizing view of future performance, but in the process raises further questions on just how AMD is doing it.
In any case, however AMD is doing it, the updated geometry engines will also feature one more advancement, which AMD is calling the primitive shader. A new shader stage that runs in place of the usual vertex and geometry shader path, the primitive shader allows for the high speed discarding of hidden/unnecessary primitives. Along with improving the total primitive rate, discarding primitives is the next best way to improve overall geometry performance, especially as game geometry gets increasingly fine, and very small, overdrawn triangles risk choking the GPU.
From what they said in techreport it was designed listening complaints of console dev using GCN at low level
Last edited:
I found out something interesting: a UPS won't stop a PS4 from freaking out in a loss of power. I have a PC, modem, router and PS4 Pro plugged into my UPS, mostly because the PS4 is so twitchy about power. I lost power for an hour today and everything is still on except the PS4 Pro.
Hmm...
I didn't plug my PS4 classic to UPS but my electricity often goes out for a fraction of second. As indicated by lights that goes off for a fraction of second.
The PS4 still works just fine like nothing happened.
Maybe your UPS response time is simply too long for Pro? Or maybe Pro is more sensitive to electricity lag than PS4 classic.
I have both actually and they are both sensitive. I've found a couple other references online. So I'm not the only one.
ProspectorPete
Regular
Thanks for you extensive work goonergaz!
This SSHD would mostly improve open world titles like Forza Horizon or GTA; most other games like TR are played front to back (usually). It would help if you have kids, for single player titles at least. Make them finish the whole game before you yourself play it, and it's fast loading times
This SSHD would mostly improve open world titles like Forza Horizon or GTA; most other games like TR are played front to back (usually). It would help if you have kids, for single player titles at least. Make them finish the whole game before you yourself play it, and it's fast loading times
http://www.neogaf.com/forum/showpost.php?p=234190249&postcount=114
hm... doubled ROPs?
Thought I recall @3dilettante wondering about this with respect to "shutting off half the GPU" as the RBEs were tied to Shader Engines?
hm... doubled ROPs?
Thought I recall @3dilettante wondering about this with respect to "shutting off half the GPU" as the RBEs were tied to Shader Engines?
Well, half the GPU would still have to leave them with 32 ROPs ...
That is one scenario I was considering.
Not being "quite right" may not be the same as doubled standard ROP fillrate.
Another idea I had was something where the Pro saves on hardware by having back-end blocks that are 32 ROPs in compatibility mode, but a shift to Pro mode leaves the classical ROP throughput at 32. The other half become the path for the parallel handling of the ID buffer and other customizations.
I don't think there's been a public timing comparison for ROP-heavy operations in a game between the PS4/similar GCN GPU and the PS4 Pro to be sure.
Another possibility, if the Pro is really all-in on checkerboard, is that it's not 2x color relative to the PS4. The ID buffer and the expected higher ratio of depth to color for checkerboard may mean there's more "other" fill rate.
Perhaps a reason why the Boost Mode is limited to just the clock increase as opposed to unlocking all shader engines & funky ROPs? Or would there be some oddities with compute scheduling/queues for non-Pro titles?
hm... so "pure" depth blocks on the 2nd half (no colour)?
Last edited:
That's a potential reason if there is an asymmetry between the compatibility half and the half turned on in Pro mode. The first half would have enough hardware for a full complement of RBEs, with half of them sharing data paths or hardware with the ID buffer in Pro mode.
Turning both halves on would make each side have 16 active ROPs with the second half lacking a reason for more than 16 fully-featured ROPs.
I thought that half the others CUs were unavailable because most PS4 games were using the existing shaders is a specific way (because it's more efficient ?). But that the boost mode should theoretically unlock the others CU on games using the CUs in a general 'unified' way.
IRC there are some games that showed impressive framerate improvement in some passages that couldn't be explained by the 31% CPU boost alone, particularly Battlefield 4:
Battlefield 4: >100% framerate improvement in some cases during conquest mode
Project cars: 38% framerate improvement during rain
The talos principle: often 35% framerate improvement
- Status
