That's not been my experience. I certainly have seen my 1TB iPad use more than 4GB of RAM. But I only tracked it when using heavier single apps like Affinity, Pixelmator, etc. It could be that Safari only sees the first 4GB.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
iPad Pro How is memory management on new iPad Pro?
- Thread starter Rogifan
- Start date
- Sort by reaction score
No Ipad pro? The OG ipad pro came with 4GB in 2015 when only the air 2 had 2 and all other ipads had 1GB. So that was a huge leap back then. The big mistake was the 2GB in the ipad pro 9.7. Having said that, they haven't moved from there for almost 5 years, and the 2018 one should have come with 6GB standard on every model and the 2020 with 8GB. The superiority of RAM management in IOS is "relative". Compared to Android it does a better according to tests, but compared to a desktop system, which is the true competitor for Ipads since Android is basically a phone OS, the RAM management is just different. IpadOS relies more on compression and less on paging, while a Desktop OS does the opposite. This way even a 2GB ipad feels fast, contrary to a 2GB windows 10 device for instance, but you pay for that speed as the ipad will refresh much more, with the risk of losing data, while a desktop OS will keep much more in memory thanks to paging (to disk). If ipads want to become more similar to desktops, they need more RAM. This, like the introduction of touchpad and mouse support, will also encourage developers to port their pro software to the app store... And Apple has every interest to have that 30% cut they mostly don't get on MacOS, let alone Windows...
As bad? I have the 1 gen iPad Air and it has better ram management than my iPhone with iOS 13 lol
So it isn’t bad at all, but yes, I hope they solve that on iOS 14
Trust me even compared to Android in real usage iPadOS does not do better. I have Android phone with 4 GB RAM and I have 2018 iPP wiht 4GB RAM. Guess where I have less apps and tabs refresh. On the Android phone. And even if apps/tabs have to be reloaded it is quite faster on my phone. So yeah iOS RAM management might have been better in the past compared to Android but not anymore. This is a myth now. iOS as a whole does not have good RAM management. I have 3 Apple devices (regular iPad with 2 GB RAM, iPP with 4 GB RAM, iPhone 8 with 2 GB RAM) and all of them suffer of this sickness. Of course the iPP is better but still not what I get from my Android phone. I have had Android (granted all Sony devices) phones all of my life - current with 4 GB RAM, old with 2 GB RAM and I have never experienced such massive reloads there. Not sure who decided to proclaim iOS management as good and when they decided to do it but I have used Apple devices for 2 years now and it is not.
To be honest I prefer to not lose my data and lose a bit of speed than vice versa. Either way the speed on iPP is again only on benchmarks. My Windows laptop despite getting half of the score of the iPP is much faster and more reliable. So I am not sure that even the claims about the speed are true. Seen on benchmarks - yes. Useful when you take pictures and the device uses machine learning - most probably. Seen every day life - definitely not for me.
So I would say Apple need to get their sh*t together and both improve iOS management software wise (because really my Sony phone can do it with 4 GB RAM) but also increase RAM memory if they cannot fix the software itself. They need to choose. If they want us to see the iPP at some point as our main computing device, the RAM management needs huge overhaul and revamp. Like from ground zero.
Regarding speed, the problem with benchmarks is that people look at multicore score while what translates into perceived speed is single core scores. So my guess is that your Windows laptop has much lower multicore score and much closer single core score... People should just stop giving this much importance to multicore, especially on ipads. The air 3 / mini 5 have exactly the same speed as the ipad pro, as they have the same single core speed. Only a few apps take advantage of that massive multicore speed..
I disagree. In my experience, even for mainstream business type usage, multi-core speed is more important than single-core. Yes single-core is important, but multi-core is more so.
I think the best example of this is my 2.9 GHz triple core Athlon II to 2.8 GHz six-core Phenom II upgrade. Same single-core performance but twice the multi-core performance. Nothing else changed (same OS version, motherboard, SSD, RAM, GPU) but the machine is now noticeably more responsive even though all I do with it is check email, run MS Office, VPN to work, and surf on it (and not at the same time), with the very occasional Netflix 1080p. In modern times, the Athlon II sometimes struggled a bit, but all the pauses and delays that Athlon II had are simply gone with the Phenom II. In fact, I’d say it’s the third best computing upgrade I’ve ever done. (The best was going from a Celeron 366 MHz, to a Celeron 800 MHz, to a Celeron 1.4 GHz, all on the same motherboard. Tied for best was upgrading my Cube from a 450 MHz G4 to a 1.7 GHz G4.)
Similarly, our iPad Pro 10.5 (A10X) is faster than our iPad 7 (A10) even just for surfing. The iPad 7 is very decent, but the iPad Pro 10.5 just flies. Both are on the same version of iPadOS.
Last edited:
Here are the benchmarks I could find for the two devices I am comparing (2018 iPP and my 2017 Asus laptop):
iPP single core score - 5105
iPP multi core score - 16724
Laptop's CPU (i7-7700HQ) singe core score - 876
Laptop's CPU (i7-7700HQ) singe core score - 3303
As you can see based on benchmarks the iPP should be far faster than my laptop but this is not the case at all. Then again my usage is basic in terms of CPU on both the laptop and the iPad - it is about browsing. My usage is more like a killer for RAM because I tend to use a lot of apps and tabs in the same time. The only taxing thing I do on my laptop is photos processing in batches. But that I cannot test on an iPad because the application I am using (DxO) is not available on iOS. Plus with this corona virus I have not taken RAW pictures to process.
The other use case I have is Software Development - I either play around with Java and currently SpringBoot or with python. I do not have the setup for this on an iPad as well.
So maybe it is more about me not having the possibility to compare properly.
Those benchmark results make no sense. You must be using the wrong tools. If you are using Geekbench you are using the 4 version on your IPP and the 5 version on you laptop. The scores are completely different on those 2 versions.
[automerge]1588761346[/automerge]
We'll agree to disagree. Having said that on Windows multicore is more important than on IPadOS as many more things are happening at the same time on Windows. I don't have any A10 devices to compare, but can tell you that the ipad pro 11 and the ipad mini 5 are exactly the same speed in normal use
I was using the scores posted on the site but you are right. Most probably iPP scores were with the 5th version and the laptop's CPU scores were with the 4th version. Or vice versa. You get my point.
Indeed most probably we do more multitasking on Windows than on iPadOS just because the devices are done this way. It is easy for me to do more things at once on the laptop (and the bigger screen) than on the iPad.
Yes, multitasking benefits more from multi-core and all windows versions after XP have been optimized for multicore systems. Many windows and mac software themselves have been optimized to take advantage of multicore... Much less on ipad where most apps are not and only when used in multitasking they benefit from more cores, but even multitasking is limited on ipados.
As for geekbench the IPP scores 5000 on GB4 and around 1000 on GB5 which is very close as single core score. Multicore is almost 5000 on GB5 so quite a bit more than on you quad core laptop, but only a few apps can take advantage of that...
Definitely.
Got it. Yeah maybe apps like say Lumafusion and you editing long 4K video. But this is not my case. I mean I do not do video processing that much to be honest. Lately I use iMovies to do like slideshow of photos I have taken during a trip but that's it.
Let's not confuse your personal anecdotal experience with facts.
I don't disagree that you have issues, however my aging iPad Pro 9.7" and iPhone7+ seem to be still remarkably snappy for me.
But I am not talking about being snappy. They are not slow, they are slower compared to my Android phone. This does not make them unbelievably slow. It makes them just slower. Nevertheless the difference in the speed is not huge - I am talking about a second or two difference here.
The bigger and noticeable difference is the reloading. That just happens far more often on the iOS devices I have compared to the Android device I have. The reason the speed is mentioned is because there is claim that iOS suspends apps/tabs (hence the reloading) to give better speed. And I am saying that it does not. They devices are not snappier than my Android phone, it is relatively the same (to be honest a bit slower) and have more reloads. So in the end I don't see the gain for me as an user here.
Again, this is anecdotal evidence. This is not a fact. Like I said, I'm not disputing that you are noticing issues, but the phrasing of your post was as if to say it's a fact that iOS memory management is now worse than Android.
Your personal experience does not a fact make.
It is worse compared to my specific Android phone. I have three iOS devices and for all of them RAM management sucks. If it was one device I would agree with you but I have 3 (iPhone, 2028 6th gen iPad, 2018 iPP). I do not accept the fact that all those three devices (bought in different times and one of them from completely different country/store) are defective. This is not the case. What happens here is that for my usage (which is the same on all of my devices) iOS RAM management sucks. It sucked on iOS 12 and it sucks on iOS 13 as well. I used iOS 11 only for a day on the 2018 regular iPad so I cannot tell you whether it sucked there.
I was never impressed with iOS ram management for the 1.5 years I have been using iOS devices. It was never impressive. The amount of reloading and apps crashing is just too much to accept it as normal behavior.
Last edited:
And again, not disputing that you have all these problems. But these are not facts, they're personally witnessed anecdotal situations.
Also I'm not claiming that iOS memory management is better than Android (or worse, or on par).
All I'm saying here is that personal experience does not a fact make. That's all.
Facts are based on repeatable tests proving conclusively that a theory is true. Your experience is not repeatable.
The iPad Pro 3rd generation and the iPad Pro 2nd generation feel very similar in normal light use like general surfing because they are both already very fast.
Yes, if I look for it, I can see the 3rd gen outperforming the 2nd gen, but overall for surfing and such they are actually pretty close IMO. It’s a matter of diminishing returns.
However, the iPad 7 feels slower than both.
iPad 7 and iPad Pro 2nd gen have similar single core speed. However, for both single and multi-core speed, 3rd gen iPad Pro blows the 2nd gen out of the water in benches. Yet for actual surfing feel the 2nd gen is already great. It’s only the iPad 7 that stands out here for being slower.
Thus, comparing iPad mini 5 vs iPad Pro 11 is not so useful in this context to test your theory, because A12 is already fast enough that going to a faster A12X SoC doesn’t help much for light usage like surfing in 2020. It may matter after several years though when surfing becomes even more CPU intensive, just like it matters now in 2020 with A10 vs A10X.
Last edited:
I 100% agree on this. I no longer have the 11in but I still have the mini 5 and the 10.5 and they are pretty close. As you said, if I look for it, the mini 5 is faster, but it's negligible at this level. Not so much against the 1st gen pro, there you can see the difference more clearly. And I also think it's not just a matter of CPU but also storage speed. I had found a app to measure storage speed and I remember for instance the air 2 having much lower storage speed than the pro, which helped explain some of the difference in perceived speed. I have no idea of storage speed of the ipad 7 though...
Yep, based on my own usage.
Well to be honest I have not seen any meaningful tests in that direction. Most people talk anecdotal anyway basing their claims on rather old facts. I have been using Android phones since 2008. The truth is that Android ram management really did suck initially. Well not just initially. It got to something better in say Android 5 and then from Android 6 it is actually quite good. Android improved a lot with 6th version.
On the other hand I would not be surprised if iOS initially had way better RAM management compared to now. There is a thread both in the iPad section and the iOS section where people do complain about issues with RAM management in iOS 12 and iOS 13. Now maybe people that have been using iOS devices for years do not notice the changes from version that much.
I am a new user so I don't have the experience of the old versions. I can base my opinion only on the newer versions.
Anyway back to the point. Most claims about iOS ram management IMO are based on the past and not the current usage. I am not saying that it sucks big time. I am saying there is enough proof that something went wrong with iOS 12 and especially iOS 13. As person like me has seen only those two versions it is not surprise that I am not super impressed.
If you look at what I wrote yesterday, IOS (13) can do miracles with RAM management, so it's not that it can't, but there are some algorithms that reload things at some point, even if it's not necessary... But you can do some heavy multitasking with no reloads for some time with 2GB devices, problem is it's not always like this and it's not easy to understand what triggers Apple's criteria for reloading...
Could be but as an end user we expect things to work and to not need to know how exactly to use them to avoid the bugs. In general I am picky user that is against automation if the automation is doing things that do not work for me and I have no control over that. So I feel strongly about things like that. IMO automation makes sense and it is good when you don't notice it. I mean it is so good that you don't even realize that it exists - it makes your life easier. However the moment you notice this automation/optimization it is usually when things are not working as you want/expect them to.
As a former Software Developer/Software Team Lead/Architect now I refuse to accept any implementation about automation/optimization that does not come with perfect algorithm. As a general rule users would accept system that requires them to do something a bit more than what they want ahead of system that does more than they asked for and they have to take additional steps/time to correct the system doing. And unfortunately this is the current state of iOS ram management. It does stuff (apps reloading/tabs reloading) that lead to data being lost and I have to enter it again which basically means time lost.
Absolutely, if it was not clear, I said that as a criticism for Apple. It could be much better if they did not mess so much with RAM management... So technically it could be much improved with an update but I am not confident it will happen. And honestly Ipad will never truly replace laptops for professional use, even with pro apps, until people can be certain they will not lose their data (as it happens for instance when they come back to their ipad after a while...)
This is my concern to be honest. That while I and you see this as a problem, Apple does not it see it as a problem. If they do not treat this as a problem they would not fix it. I assume this is why some people (including me) are so vocal about those issues. We do want Apple to acknowledge this and change it.
People will become vocal as they begin using ipad professionally more and more... Like when people rejected windows 8 and Microsoft had to go back, or more recently with forced restarts of Windows 10 that caused data losses and pushed Microsoft to give even Home users a way to pause updates. So it will happen when enough people complain...