A Lot of Geekbench Scores of Completely Arbitrary Devices (because I wanted to)

Matt Birchler | | 3 min read

This is going to be a pretty simple post, as my wife got a new laptop this month and I wanted to see how it and some of the other devices around the house. If nothing else, this gives you an idea of how far we’ve come, and how amazing the chips are in the phones we carry around in our pockets.

The Devices

  1. Dell XPS 13 7390 2-in-1 (10th Gen Intel Core i7-1065G7)
  2. MacBook 2016 (1.1 GHz dual-core Intel Core m3-6Y30)
  3. iPad Pro 2018 (A12X)
  4. iPhone 11 Pro (A13)
  5. Pixel 4 (Snapdragon 855)
  6. Mac Mini 2012 (2.5 GHz Dual-Core Intel Core i5)

Obviously, these devices span 7 years and have wildly different prices, so this is more a curiosity than a “who will win!?” head-to-head.

Geekbench Single Core

The latest iPhone gets the best single core score here, but the Dell XPS laptop with a 10th gen i7 is right there, as is the iPad Pro from 18 months ago.

Geekbench Multi-Core

Things change a bit when we switch to mili-core performance. The XPS edges out the iPad Pro by 0.2%, which is basically a tie, and the iPhone 11 Pro drops down to a distant third. Meanwhile, the Pixel 4 separates itself a bit from the MacBook and aging Mac Mini, both of which are far and away the slowest.

Geekbench Compute

And finally we have the compute test which test graphics performance, and in this case the iPad Pro shot up to 13% faster than the XPS, and the iPhone 11 Pro maintained a similar lag compared to the two laptop-class devices. The 8 year old Mac Mini also fell off a cliff here, to the point where I wondered if it would even finish the test at a certain point.

Price Per Point

This is far from a perfect measure and it does not indicate the intrinsic value of each device, but I thought it was interesting to compare the price of each device and calculate how many points of multi-core performance you get from each device.

By this imperfect measure, the MacBook was far and away the worst value and the iPad Pro was the best.

But of course, this all depends on the measure you use. Here’s the same chart based on the single core performance:

Now the iPhone is the winner.

And just to complete the set, here’s the price precompute point:

I think the one thing this tells us for sure is that no matter how you slice it, the 2018 iPad Pro is a great deal and is a little powerhouse 😉

Most 2-in-1 Laptops Haven’t Gotten That Much Better

Matt Birchler on link | | 2 min read
Most 2-in-1 Laptops Haven’t Gotten That Much Better

Finding Inspiration in the Microsoft Store — MacSparky

A few years ago I played with some of the initial 2-in-1 computers and they felt a lot more like the worst of both worlds than the best at the time. The 2-in-1s I saw on my recent visit were much improved. The devices were lighter, the screens were brighter, and some of them had plenty of power to get most computing done.

This is a sneak peek at an article I'll be writing soon, but my wife upgraded her laptop this month. She went from a 2015 MacBook to a 13” Dell XPS, a 2-in-1 laptop. The screen flips around and you can technically use it as a tablet as well, but I wouldn't recommend it.

As David mentions, these devices used to be the worst of both worlds, and that is no longer the case. The XPS is a fantastic laptop, and when used in that form factor it's really excellent. But when you spin the screen around to throw it in tablet mode, it all falls apart for me. Not only is Windows a major pain to use in tablet mode, but imagine having a tablet that weighed 3 pounds and was 0.5 inches thick (aka 2x as heavy and 2x as thick as the iPad Pro). And while the keyboard on the back on the device is annoying on the iPad Pro as well, you can take the keyboard off the iPad, but it's locked in on the XPS.

In my opinion, the stuff Microsoft is doing with the Surface line is the most compelling 2-in-1 work on the market today. I look at products like the Surface Book 2 and think "yes, that's it!" Take any other issues you have with it aside and from a purely form factor perspective, it's built to be great in laptop mode and tablet mode, and it does so by making the keyboard detachable. Us Apple-centric folks may not like Microsoft's OS as much, but if a MacBook Pro came in this configuration we would be all over it.

HomePod is Growing, but is Still Pretty Niche

Matt Birchler | | 1 min read

Strategy Analytics: HomePod sales grew in 2019, but Amazon and Google still lead - 9to5Mac

I enjoy my HomePod, I really do! It's not the best smart speaker out there, but as someone who is heavily into the Apple ecosystem, it's the best option for me.

As for why the HomePod saw some growth last quarter, I have to think the price drop to $299 helped, if for no other reason that it made the sale price over the holiday in many places $199, which I think is a perfect price for what this thing offers.

Which brings me back to the thing I've been suggesting since 2017 😬

If for no other reason than the privacy advantage of having an Apple mic in your house vs one from Amazon or Google, having a more affordable and more capable HomePod would be a great win for consumers and Apple overall.

Shipt's "Bringing the Magic" Is Insane

Matt Birchler | | 1 min read

Target's Delivery App Workers Describe a Culture of Retaliation and Fear - Vice

Shipt workers told Motherboard that customers who order from Target often seem surprised when independent contractors in plain clothes driving their personal cars show up at their homes with massive deliveries from Target.

And:

Workers say Shipt customers often live in gated and upscale communities and that the app encourages workers to tack on gifts like thank you cards, hot cocoa, flowers, and balloons onto orders (paid for out of their own pocket) and to offer to walk customer’s dogs and take out their trash, as a courtesy. Shipt calls this kind of service “Bringing the Magic,” which can improve workers’ ratings from customers that factor into the algorithm that determines who gets offered the most lucrative orders.

On the first point, I totally get the confusion. I too was confused when this rolled out because Target really talked it up as though this almost just what they were calling their own delivery service. And when you look at the app today…

It's even harder to differentiate where the Target and Shipt delivery line is.

  • Order for in-store pickup: Target
  • Order for drive up pick up: Target
  • Order for next day deliverey: Target + UPS
  • Order for normal delivery: Target + UPS
  • Order for same-day delivery: Shipt

That second bit about writing thank you notes, walking the dog, and taking out the trash is maybe the most insane thing I've heard this year though (and we have a maniac in the White House, so you know…). First off, this is absurd to ask your workers who aren't getting paid nearly enough, and second, if someone dropped off my groceries and then asked to walk my dog I'd just about want to call the police because a dog theif is on the loose.

iPhone 11 Pro, Pixel 4, and S10e Low Light Shoot Out

Matt Birchler | | 1 min read

I took out my three Android phones this week and took a bunch of photos around 10PM to see how they all did in really low light situations. Checkout the video above to see the details, but the gist is:

  1. The iPhone is the best at capturing details in the dark
  2. The Pixel 4 tends to have the best color
  3. The selfie camera on the iPhone is bad in low light
  4. They're all shockingly good compared to where we were even 2 years ago

What Even is a Video Game Speed Run (Ocarina of Time)?

Matt Birchler on link | | 2 min read

I’ve never been that into video game speed runs, but I have had an obsession with The Legend of Zelda: Ocarina of Time speed runs for years now. In 2013, a speed runner called Cosmo did a run through Ocarina of Time in 22 minutes and explained both what he was doing and what the history was around the glitches he was using to achieve that 22 minute time. If you havbn’t watched it before, it’s absolutely worth a watch.

Since then, the best time has dropped time and again to just under 10 minutes. Considering the game proper takes 20+ hours to beat normally, this was really insane.

The below video explores the changes in the Ocarina of Time speed run methods and whether or not speed running this game has shifted into something that’s sucked all the fun out of Ocarina of Time runs.

It’s a complicated question, and I’m personally conflicted. On the one hand, the method shown off in 2013 and the one that was used to get down to 17 minutes in 2019 used glitches too, so I don’t think that using glitches in the game to skip to the end of the game is fundamentally wrong.

But on the other hand, the 17 minute runs were way more fun to watch. They took you outside of the opening area of the game, they involved a couple boss fights, they partially relied on getting lucky with some randomly falling debris at one point, and maybe most of all, it was easy to understand what they were trying to do to get the glitch moments to happen. This new 10 minute method lacks all of this, and basically amounts to watching someone seemingly walk randomly around a small area until they’re warped into the end credits.

All I can say for sure is that for me, this new method for beating Ocarina of Time in next to no time is of zero interest to me as a casual viewer. The drama of the old style of speed run is what kept me coming back whenever a new record time was posted. Would they pull off the jumping combo on the first try? Would they beat the boss in an optimum time? Would they just get hit with a random bolder when running up the castle? With all of this gone and the run being reduced to a bunch of random side-stepping and looking around at nothing, the thrill just isn’t there for me.

The Samsung Galaxy S20 Lineup

Matt Birchler on samsung | | 1 min read

In 2019 the new flagship iPhone lineup was:

  • iPhone 11: $699
  • iPhone 11 Pro: $999
  • iPhone 11 Pro Max: $1,099

And the 2019 Samsung Galaxy S10 lineup was:

  • Galaxy S10e: $749
  • Galaxy S10: $899
  • Galaxy S10+: $999

Pretty darn similar, right? Well, here is is the new Galaxy S20 lineup.

  • Galaxy S20: $999
  • Galaxy S20+: $1,199
  • Galaxy S20 Ultra: $1,399

And if you throw in the Galaxy Z Flip, that starts at $1,380.

This is a lot of numbers, but it’s interesting to see Samsung completely abandon the sub-$999 market for their flagship lineup. While Apple had the $699 iPhone 11 as their top seller every week since it was released last September, Samsung eliminated their lower cost option and is all in on not only high end, but higher than Apple high end pricing. Meanwhile, we have rumors of an impending iPhone SE 2 coming from Apple that will bring top-tier specs in a $399 price point.

The S20 lineup looks great, and Samsung has crammed these phones with high-end everything, but this pricing is really a surprise to me. Apple’s top selling flagship is $50 less than last year’s model and they have a great small phone waiting in the wings. Meanwhile, it’s Samsung who’s going all in on the ultra-high end.

HDR Testing Halide, Lightroom, and Stock Camera App on the iPhone 11 Pro

Matt Birchler on camera, iphone | | 3 min read
HDR Testing Halide, Lightroom, and Stock Camera App on the iPhone 11 Pro

As I’m one to do, I saw a cool lighting situation this morning and decided it was a great chance to do a photo comparison. I only had my iPhone with me, so I decided to test how well the three apps I use the most for photography handle HDR situations. First, a quick recap of each app.

Halide

An amazing app for RAW image capture. As far as I know, this app does nothing special for HDR situations, but delivers a high quality RAW (DNG) image file that you can use to edit later.

Lightroom

Mostly used for editing, the camera component of Adobe Lightroom does not get the love it deserves. You can toggle the app into an HDR mode, which will capture several images, patch them together, and do some automatic edits to save your highs and lows.

Stock Camera

This is what most people use, and as of the last year or two, it’s my go-to for most mobile photography as well. It takes upwards of a dozen photos and combines them into one shot that is just a JPEG, but should have the highs and lows fixed so I can see them clearly.

Shadows

Let’s look at the dark part of the image first:

Halide

Lightroom

Stock Camera

I think Halide is the worst of the bunch here, both in terms of retaining details, as well as in image clarity. Of the two remaining, I think I have to go with the stock camera app due to its excellent clarity, lack of noise, and generally good data retention. Lightroom captures more information in the tree, but there are enough artifacts from the multi-image capture (see the top of the tree) to make it not worth the extra detail.

Highlights

Next, up let’s look at the highlights.

Halide

Lightroom

Stock Camera

Again, Halide doesn’t seem to be built to handle this as well (I can get a little more detail by cranking down the exposure of the image, but then everything else looks like trash), so this is a two horse race again. This is pretty close, because the stock camera app gets more separation between the sun and clouds, but Lightroom has much better looking clouds and lacks the overly orange color around the sun. I’m going to consider this one a personal taste issue, so whichever you prefer is the winner.

Conclusion

In this specific shot, I think the iPhone’s stock camera app does the best job of producing the best image with the most acceptable downsides. Yes, the sun has some coloring issues, but I fixed those in (ironically) Lightroom in 5 seconds. Yes, the clouds are overly “painterly” as the machine learning didn’t love the wispiness of the clouds, but I can live with that since you only really notice it when zooming in.

Overall, the stock app continues to impress me with its ability to get great photos in many situations all without making me think about much other than composition, which is excellent. Oh, and while Lightroom, which I love as well, takes about 1 second to capture it’s multiple images in its HDR mode, the stock app does it effectively instantly.