Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

The "new" PS5 actually has better cooling!

Summary

 

After Austin Evan's famous video about the new PS5 revision, the CFI-1116A model, it turned out that it had a smaller heatsink that weighted ~300 grams less. A debate started on the whether Sony *cheapened out* on cooling performance or not. Ares, a Greek electrical engineer who has his own small tech channel called "Hardware Busters" decided to do a deep test into it and find out the truth. As it turns out not only the new heatsink offers good cooling it even offers 11 decrees lower in the SoC.

However even with that cooling difference, it doesn't really matter since the both models get the same power of the wall and have the same SoC thus get the same performance.

Quotes (translated from Greek by yours truly)

Quote

Despite what has been said lately, in our measurements the new PS5 seems to be somewhat better than the original model, at least the model we had at our disposal. Taking into account that the exhaust temperature on the new PS5 is higher than the "old", it means that the cooling system does a better job.

My thoughts

Well, Sony engineers are actually good at their job. Sadly Ares isn't an expert when it comes to heatsinks so he can only observe and document the results. I hope that someone who is, eventually comes around to answer the new question in my mind. "Why is the smaller heatsink better?"

 

Sources
Link to Austin Evan's video:

Link to Ares' written news article, warning it is in Greek: https://www.insomnia.gr/review/video-games/sony-playstation-5-cfi-1116a-review/ But the Temperature and Voltage readings as well the Thermal Camera pictures are there.
Usually he also writes for some popular English tech news sites, but if he has in this scenario I wasn't able to to find it. Anyway for those that want to understand his commentary and input on the situation here is the
Link to the youtube video on their English channel "Hardware Busters Internationals", It is basically the news article but in an English video form.

 

Link to post
Share on other sites

I think it has more to do with optimization of the cooler to better follow the path of airflow here rather than anything else. The removal of a piece of the heatsink was most likely done to remove restrictions and because the air would be hot anyway thus the cooling potential of the other parts of the heatsink would be lowered as the difference in temperature would be lower.

Link to post
Share on other sites

imo they were blindsided by the launch of the series x/s so they threw a console together as fast as they could in order to meet a certain time frame, so now theyre going back and optimizing that design

Link to post
Share on other sites
33 minutes ago, Katarok said:

Well, Sony engineers are actually good at their job. Sadly Ares isn't an expert when it comes to heatsinks so he can only observe and document the results. I hope that someone who is, eventually comes around to answer the new question in my mind. "Why is the smaller heatsink better?"

Law of Diminishing Returns is daddy in heatsink design.

Remember to press quote to get a response from someone! | Check people's edited comments! Press F5 to do so! | Remember to press that check mark on the right answer that solve your problem! | React to your heart, not peer pressure!

 

"Wow, you want me to say ara ara? Really?"

- Clock Girl

Self Introduction and Current Rig

 

🏳️‍⚧️ Rights

Link to post
Share on other sites

Well I can tell you it's lighter because there's less COPPER in the heatsink, and there's less aluminum fins. 

They're countering that by changing the fan blade, the fan has longer blades, which to me feels like  optimizing for more air pressure, and more air gets actually pushed through the fins. 

 

To me an uneducated guess would be that the engineers were probably overly concerned about dust blocking the heatsink and causing hardware failures so they oversized the heatsink initially, and now they realized it's not an issue so they're reducing the heatsink a bit. 

 

It could also be they're preparing for a new batch of chipsets or ps5 chips made at 6nm instead of 7nm, or the "refined" 7nm process from tsmc, which would cause a tiny reduction in heat produced by the chips .. I doubt it could be this, but wouldn't be impossible.

 

(I may be wrong about the fan blade optimization but that's what seems logical to me). 

 

Link to post
Share on other sites
Quote

After Austin Evan's famous video about the new PS5 revision, the CFI-1116A model, it turned out that it had a smaller heatsink that weighted ~300 grams less. A debate started on the whether Sony *cheapened out* on cooling

People are so ill-informed. 

 

Weight of a heat sink tells us nothing of it's cooling capacity other than it takes longer for it to reach equilibrium temperature because of the increased thermal mass. 

Link to post
Share on other sites
3 hours ago, Katarok said:

However even with that cooling difference, it doesn't really matter since the both models get the same power of the wall and have the same SoC thus get the same performance.

This is actually untrue. Better cooling will allow different boost behavior in either the GPU or CPU portion of the SoC. even if the maximum power envelope remains unchanged as the chips' max clocks are temp dependent. Probably not going to be that much of an issue in practice, but it will effect performance on the margins.

Link to post
Share on other sites
3 hours ago, ki8aras said:

imo they were blindsided by the launch of the series x/s so they threw a console together as fast as they could in order to meet a certain time frame, so now theyre going back and optimizing that design

This doesn't really jive with PlayStation history.  The PlayStation's have always been constantly revised beyond the major revisions that consumers see such as 'Fat' and 'Slim' models.  The PS1 had around 12 major motherboard revisions.  The PS2 has over 20.  The PS3 has 20 or so.  While the PS4 had less, the basic PS4 had about 6 revisions and the PS4 Pro had 2-3.

Sony going back and optimizing the design is not a result of 'rushed development' it's literally what they've always done with their hardware for the last 26 years.

Desktop: Ryzen 9 3950X, Asus TUF Gaming X570-Plus, 64GB DDR4, MSI RTX 3080 Gaming X Trio, Creative Sound Blaster AE-7

Gaming PC #2: Ryzen 9 3900X, Asus TUF Gaming B450M-Plus, 32GB DDR4, Gigabyte Windforce GTX 1080

Gaming PC #3: Intel i7 4790, Asus B85M-G, 16B DDR3, XFX Radeon R9 390X DD XXX OC

UnRAID #1: Intel E5-2697v2, Asus P9X79 LE, 48GB DDR3, Radeon HD 5450

UnRAID #2: Intel E5-2603v2, Asus P9X79 LE, 24GB DDR3, Radeon HD 5450

MiniPC: Lenovo M93P Tiny: i7 4950HQ, 16GB DDR3 
Windows XP Retro PC: Intel i3 3250, Asus P8B75-M LX, 8GB DDR3, Sapphire Radeon HD 6850, Creative Sound Blaster Audigy

Windows 9X Retro PC: Intel E5800, ASRock 775i65G r2.0, 1GB DDR1, AGP Sapphire Radeon X800 Pro, Creative Sound Blaster Live!

Link to post
Share on other sites
9 minutes ago, ravenshrike said:

This is actually untrue. Better cooling will allow different boost behavior in either the GPU or CPU portion of the SoC. even if the maximum power envelope remains unchanged as the chips' max clocks are temp dependent. Probably not going to be that much of an issue in practice, but it will effect performance on the margins.

Only if you reach the temperature threshold.

 

If it is the case with both heat sinks in the PS5 that you never reach the temperature threshold there shouldn't be any difference in performance. 

Link to post
Share on other sites
17 minutes ago, Spindel said:

Only if you reach the temperature threshold.

 

Temp threshold for AMD CPUs/GPUs is a soft threshold, not a hard one. 

Link to post
Share on other sites
57 minutes ago, CerealExperimentsLain said:

Sony going back and optimizing the design is not a result of 'rushed development' it's literally what they've always done with their hardware for the last 26 years.

Not limited to Sony. Many products will not be the best they can be to get it out within a certain timeframe. This is fine as long as it does what it is supposed to. After launch, they can then refine it where it is worth doing so, usually motivated by cost savings, or if required for other reasons e.g. changing part availability.

TV Gaming system: Asus B560M-A, i7-11700k, Scythe Fuma 2, Corsair Vengeance Pro RGB 3200@21334 4x16GB, MSI 3070 Gaming Trio X, EVGA Supernova G2L 850W, Anidees Ai Crystal, Samsung 980 Pro 2TB, LG OLED55B9PLA 4k120 G-Sync Compatible
Streaming system: Asus X299 TUF mark 2, i9-7920X, Noctua D15, Corsair Vengeance LPX RGB 3000 8x8GB, Gigabyte 2070, Corsair HX1000i, GameMax Abyss, Samsung 970 Evo 500GB, Crucial BX500 1TB, BenQ XL2411 1080p144 + HP LP2475w 1200p60
Desktop Gaming system (to be retired): Asrock Z370 Pro4, i7-8086k, Noctua D15, G.Skill Ripjaws V 3200 2x8GB, Asus Strix 1080 Ti, NZXT E850 PSU, Cooler Master MasterBox 5, Optane 900p 280GB, Crucial MX200 1TB, Sandisk 960GB, Acer Predator XB241YU 1440p144 G-sync

Former Main system (to be retired): Asus Maximus VIII Hero, i7-6700k, Noctua D14, G.Skill Ripjaws 4 3333@2133 4x4GB, GTX 1650, Corsair HX750i, In Win 303 NVIDIA, Samsung SM951 512GB, WD Blue 1TB, Acer RT280k 4k60 FreeSync [link]
Gaming laptop: Lenovo Legion, 5800H, DDR4 3200C22 2x8GB, RTX 3070, 512 GB SSD, 165 Hz IPS panel


 

Link to post
Share on other sites
2 hours ago, Spindel said:

Only if you reach the temperature threshold.

 

If it is the case with both heat sinks in the PS5 that you never reach the temperature threshold there shouldn't be any difference in performance. 

It totally depends on how the boost is setup. Most modern gpus have the boost being heavily tied to temperature. I mean if gpus didn't start down clocking before hitting a critical temperature threshold then that would be bad. I would imagine that the ps5 has something similar as it would likely be the easiest way to avoid having thermal shutdowns and other issues related to improper temperature control. Either that or they have the gpu at a fixed boost and created an overengineered cooling system to make sure that even at fringe cases the likelihood of thermal shutdown is small. 

Link to post
Share on other sites
3 hours ago, ravenshrike said:

This is actually untrue. Better cooling will allow different boost behavior in either the GPU or CPU portion of the SoC. even if the maximum power envelope remains unchanged as the chips' max clocks are temp dependent. Probably not going to be that much of an issue in practice, but it will effect performance on the margins.

How much energy something draws is directly tied to temperature. Since that is where the heat comes from. When a system detects that it is overheating it will reduce the amount of energy it draws to lower the heat and thus the temperature. If it doesn't it will burn itself to death.

That is what drops clocks and causes the so called "Thermal Throttling". The "boost" is simply the cooling having more overhead and thus allowing more energy to enter the system, increasing the clocks.

You can detect all these things simply by examining how much energy it draws of the wall. Obviously different architectures manage electricity differently, but the PS5 versions all have the same SoC. Thus we can tell by just looking at the numbers that the performance on both models is exactly the same.

Link to post
Share on other sites

I already knew that the updated model would have better cooling, with the decreased exhaust temperature..

"We also blind small animals with cosmetics.
We do not sell cosmetics. We just blind animals."

 

"Please don't mistake us for Equifax. Those fuckers are evil"

 

This PSA brought to you by Equifacks.
PMSL

Link to post
Share on other sites
On 9/9/2021 at 2:33 AM, Katarok said:

After Austin Evan's famous video about the new PS5 revision, the CFI-1116A model, it turned out that it had a smaller heatsink that weighted ~300 grams less. A debate started on the whether Sony *cheapened out* on cooling performance or not. Ares, a Greek electrical engineer who has his own small tech channel called "Hardware Busters" decided to do a deep test into it and find out the truth. As it turns out not only the new heatsink offers good cooling it even offers 11 decrees lower in the SoC.

However even with that cooling difference, it doesn't really matter since the both models get the same power of the wall and have the same SoC thus get the same performance.

Quotes (translated from Greek by yours truly)

My thoughts

Well, Sony engineers are actually good at their job. Sadly Ares isn't an expert when it comes to heatsinks so he can only observe and document the results. I hope that someone who is, eventually comes around to answer the new question in my mind. "Why is the smaller heatsink better?"


Overview as an non-expert/amateur that took only tangentially related course work and doesn't do this for a living: 

1. Weight doesn't mean quality. Design can matter just as much if not more (and different designs work better for different airflow configurations). More, thinner fins tends to work better if you can ensure adequate airflow. Fluid dynamics are hard (computer simulations still aren't quite there) and figuring out the best way to design a heatsink so that air passes over it and removes a maximum amount of heat while still being practical is hard.

2. Longer doesn't mean better. A heat pipe can only carry heat so far. And the further a part of a fin is from a heat pipe, the harder it is for heat to move there. Most of the heat transfer is occurring relatively close to the CPU or part being cooled. 

3. It's really hard to get air to agree with a given design. It's largely guess work. Getting the air to make good contact with heatpipes/fins is hard because turbulence and laminar flow are CRAAZY complicated to simulate. You want a steady stream of cool atoms/molecules gong across your heatsink and you don't want air 'sticking' around a fin and insulating it while the bulk of molecules move between fins without removing much heat.

 

The ultimate goal of a heatsink is to do a good job of moving the required amount heat away in a reasonable timespan at a reasonable price. 

 

Larger temperature gradients mean faster heat transfer - doubling the temperature difference between a hot and cold part doubles how quickly heat tries to move from hot to cold (all else equal). Similarly, longer distances for heat to travel equates to slower temperature transitions from one end to the other. Thus, if there's not much of a temperature gradient and a long distance to span, adding on MORE material to the far end of the heatsink hardly matters (think taller/longer heatsinks). There are SOME benefits to having more thermal mass in a design though - namely that it takes longer for the heatsink to heat up fully. This makes for more stable temperatures - temperature swings on a processor can be just as bad as high temperatures - or allows for a larger thermal load to be handled (but doesn't help much with smaller thermal loads).


As stated, it really is a game of diminishing returns. The faster you remove heat (making the heat source cooler), the harder it is to remove additional heat. Also the "best" materials might be hard to machine (it's easier/cheaper to use a lot of thin slices of aluminum - getting thin slices with copper is hard) or stupidly expensive (pure silver heatsinks aren't worth it - silver is around 9% more thermally conductive than copper but around 100x more expensive). 
 

 

This is the general idea behind A LOT of cooling. - https://en.wikipedia.org/wiki/Newton's_law_of_cooling

 

R9 3900x | 64GB RAM | RTX 2080

1.5TB Optane P4800x | 2TB Micron 1100 SSD
HD800 | SR6011 + 4xB1 + PB12-nsb
QN90A | Topre Realforce Keyboard

Link to post
Share on other sites

saw some of the most recent steve/GN's video.

a bit mixed in temps, but overall better performance in temps for the cooling.

but still overall concerns. still having a set of heating issues. also airflow differences it can have between the metal and fan.

 

also any change in software/firmware?

 

Link to post
Share on other sites
On 9/9/2021 at 5:41 AM, ki8aras said:

imo they were blindsided by the launch of the series x/s so they threw a console together as fast as they could in order to meet a certain time frame, so now theyre going back and optimizing that design

To add to what Cereal said: that's not how console design works.

 

It takes years to get a console ready: basic engineering, securing custom components, establishing a developer framework, testing... you don't just wait for your competitor to do something and then create an entire platform in a knee-jerk reaction. It's not like Sony was going to ride the PS4's coattails forever.

 

I'm reminded of how people parroted the myth that Apple's iPhone was a copy of the LG Prada announced a few months earlier. As if Apple had no plans for a phone until it saw the Prada, and magically created an entire hardware architecture and operating system in less time than Microsoft takes to patch some Windows bugs. In both cases, people were more than a little optimistic.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×