Not to criticize or to be negative to your points, as I respect the posts you have and the effort you have been making. However, I think this thread is getting to a stale point in terms of circular arguments. That said for those who are not well informed and stumble upon this conversation, I feel it needs to be said:
The A8 in the iPhone 6, yes there were some improvements. But the A8 as a chip itself? Compared to the A7? The improvements were marginal. Your posts keep pointing to benchmarks but you must keep in mind the architecture of the chips. This refresh rate stuff you are getting from an outdated PC centric benchmark is imo mostly unsubstantiated, and I am not sure why it keeps being used. I am sure It has morphed into a benchmark standard, and too often people get caught up in the benchmarks, forgetting about the actual chip and how it works together. When we look at the chip itself, the A8 may be coined as the "next gen" chip from Apple, but the components it used were of the same "family" of conponents as the A7...they were more or less simply enhanced. The actual next generation versions of those components were used in the A9, and along with a smaller fab, that is why the A9 sees much larger gains than what the A8 saw relative to the A7.
Think of the A8 as more of a refinement of the excellent architecture used in the A7. Even in the above linked benchmarks, when considering a CPU and how it relates to every day tasks, a 12% performance improvement is not substantial. Even a 25% improvement is not substantial. An 80% improvement? Yes, that is significant - as that borders closer to completing tasks twice as fast. But when we are simply opening up a calendar app or weather or messages or phone, it is not something that is significant (I suppose an argument could be made for how every gain counts, and that is where two sides of CPU debates also exist even among the die shot lovers and chip fanatics)
The A8, in a nutshell, uses the same CPU as the A7 with some enhancements. That is it. There are a few modifications to make it more efficient, and work on a smaller process, but its essentially the same skeleton; notable architecture wise too right down to the same cache, how said cache is used, how the pipeline functions and its width...its more or less, the same. That is why they are so similar in the first place. Consider: Yes the A8 in the iPhone 6 benchmarks higher, but the A8 in the iPod Touch 6G benchmarks slightly slower (relative to the iPhone 5s's A7, or even the A7 in the iPad which runs at the same clock speed as the A8 in the iPhone 6) in most CPU benchmarks, and, actually in some GPU benchmarks the touch 6g is either on par; slightly lower or slightly higher as well...yet it uses the same A8 chip clocked at a ~300MHz slower speed than the 6. The point I'm making is that they A7 and A8 are more on "par" than any of Apples previous chips, save the original iPhone / 3G or maybe even the 3GS and 4 (but at least the 4 has 2x ram...and that itself is a completely different story and debate for another time). But why the A6 in the 5 was clocked @ same speed iirc as the A7 in 5s, yet the results were completely different. A7 was in a different category of mobile processor. (And I mean relative to performance, not to it being a 64 bit chip)
If we stick to benchmarking apps, then yes there are multiple GPU benchmarks that show a solid improvement - as is to be expected but then again few apps/games push the A7 GPU as it is, to the breaking point where one needs a new GPU to handle said app/game. And even then, the gains the A8 made are pointless today, and pointless to argue over, when the current high end chip Apple has to offer is the A9 which is a pretty large leap over the A7. So when people clamour for the A9 in this new 4" iPhone, it is more understandable than having to defend the merits of the A8 over the A7. Also keep in mind, the GPU in the A8 is of the same "family" (GX variant) as the GPU in the A7 with architecture tweaks and, again, enhancements. It is like a very optimized version of the GPU in the A7. The A9 actually used a quantifiable next generation chip, and it is why we see more gains with that vs. the A8, than we do with the A8 vs A7.
(You can also look at basemark X as another benchmark indicator of GPU performance. The 6 comes out stronger when it is pushed harder, and similar to A7 when performing normal tasks or idle)
Add to that, the RAM in the 5s and 6 were identical in terms of type (lpddr3), speed (I'm pretty sure they can same freq) and capacity (1GB) and it is even less convincing hardware wise. One can further understand why 1GB in the 5se would thus be a let down in this regard.
There were significant improvements Apple made to enhance the efficiency of the chips, but at its core, the A8 was more or less an enhanced A7. We also need to keep in mind the A8 was on a 20nm process, which is a pretty big upgrade from the process the A7 was on. That can also lead to all sorts of improvements in a chip.
The general design of the A7 was enhanced with the A8, but it was not actually "evolved" until the A9. For whatever reason.
On a final note, I think the point people are making here, and it is a valid point, is that a lot of people who upgraded their iPhone 5s's to the 6, and spent a lot of money they didn't need to spend for what amounted to a small upgrade. So the thinking goes that if this "5se" is similar to the iPhone 6 internally, then it is a similar meagre upgrade. Now lets say you are someone who purchased the 5s on launch day at the end of 2013 and - for whatever reason, justified or not - you feel the need to upgrade to a new 4" device. You have waited some time for this 5se and to learn it is a similar "jump" (or rather, hop) in power compared to your current phone, one can understand where the frustration would set in
Now if you have an iPhone 5c, or an iPhone 5, or an even older iPhone, it is a much more attractive upgrade spec wise.
Hopefully it just comes with an A9, 2GB of ram, and this thread can be rendered pointless in the end.
