Archive for September 19, 2019

Thursday, September 19, 2019

APFS Enumeration Performance on Rotational Hard Drives

Mike Bombich:

My APFS-formatted rotational disks have always felt slower than when they were HFS+ formatted. The speed of copying files to them felt about the same, but slogging through folders in the Finder was taking a lot longer. At first I shrugged it off to the filesystem being new; “It just needs some tuning, it will come along.” But that performance hasn’t come along, and after running some tests and collecting a lot more data, I’m convinced that Apple made a fundamental design choice in APFS that makes its performance worse than HFS+ on rotational disks. Performance starts out at a significant deficit to HFS+ (OS X Extended) and declines linearly as you add files to the volume.

[…]

After the very first simulation, APFS starts at a deficit — APFS takes three times as long to enumerate a million files on a rotational disk compared to HFS+ enumerating the exact same collection of files on the exact same hardware. This result on its own is staggering. As you add and remove files from the volume, however, the performance continues to decline. After just 20 cycles, APFS enumeration performance is 15-20 times worse than HFS+ performance.

This seems to be because it doesn’t keep the filesystem metadata contiguous.

Previously:

Scanbot Goes Freemium

doo (tweet):

To start with the most important point: none of our users gets taken away something they have had before. That means that every feature that was unlocked by purchasing the ‘Scanbot Pro’ upgrade before the release of Scanbot 9 will stay valid and every unlocked feature will remain available.

[…]

We provide a lot of functionality for very little money: In the most extreme case, a user that bought Scanbot Pro in 2014 for as little as 0.99$ has enjoyed 9 fully featured major versions with over 400 internal builds of Scanbot.

[…]

Just additional new features that will be released from now on will only be unlocked with the subscription.

It’s $22.49/year, whereas it used to be $7 one-time. They offered me the first year for $4.49, since I’d purchased before.

Via Greg Hurrell:

The thing about a scanner app is it’s a bit like a screwdriver: you want it to do one thing, and once you have a tool that does that thing well you’re pretty much happy with it as-is. You don’t want to be locked into a a recurring payment contract just to keep screwing in screws.

The developers make the argument that a subscription model will allow them to keep working on the product. Sure it will, but it misses the point: people don’t need or want them to endlessly evolve an already-working screwdriver.

[…]

Sadly, I don’t think I’ve ever seen a small subscription based app that justified its ongoing costs: no number of pointless “bug fixes and performance improvements” releases will deliver me any perceptible value

Maybe the plan has changed since he updated, but as described it sounds like a win-win. If you don’t want the extra features, you get the basic stuff at no additional charge, forever. You can also buy certain features through one-time IAPs.

But he raises a good point about screwdriver apps. Some apps just don’t need a stream of new features, but all apps need maintenance. How do you fund that? Relying entirely on new customers doesn’t seem like a good plan. In theory, a very cheap subscription would be fair to both sides. But I haven’t seen that done. One issue is how to get your installed base to subscribe if you’ve already delivered the features they want. Another is that even if the subscription is cheap, a lot of customers will search for any non-subscription alternative. So the math no longer works, and you need a much more expensive subscription.

Previously:

Update (2019-09-27): Penbook tried the $1/year subscription.

Update (2019-10-04): Riccardo Mori:

Speaking as a customer, this drive towards subscriptions is killing my interest in looking for new apps for my iOS devices.

[…]

One of the main reasons developers constantly bring up to justify their switch to a subscription model is that subscriptions are needed to fund the continued development (or maintenance) of their apps. Okay. Do you know what more than a few regular folks have told me about this? That it sounds like a poor excuse. They have been updating the app so far without subscriptions, what exactly has become so expensive all of a sudden?

I have wondered that, too, and the answer seems to be that (for lots of apps) times have changed because growth slowed. So the older purchasers had been in effect getting free updates subsidized by newer purchasers. I also think a lot of developers were counting on Apple to eventually make paid upgrades work with the App Store. Instead, they added subscriptions.

John Voorhees:

I suspect the growth from new users was enough to keep things going for quite a while. Of course that’s not sustainable though.

Ryan Jones:

As always, John nailed it.

In 2013 it looked like growth for $3-5 apps was unbounded.