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

Curufinwe_wins

Member
  • Content Count

    10,642
  • Joined

  • Last visited

Everything posted by Curufinwe_wins

  1. As a Tesla owner with normal autopilot... literally the only feature that interests me one bit is the auto lane change, and only on big road trips (9k miles in my car in 3 months). Normal autopilot does everything I'd want it to for a regular drive. Oh and in general, it still gets spooked really easily, so yeah, if you dislike random heart-rending braking when the road dips under a bridge etc, not the tool for you.
  2. It would be fair to say that I am not an expert at these things either. Ask me about nuclear power and/or engineering materials for power systems and I'll straight up claim expert level knowledge. This? Nope. I do remember anandtech trying to break out the IO chiplet and infinity fabric consumption from first gen Zen and TR, and that was what I was basing my standards on. I agree with most of the details you say here. According to Andrei's work, Cinebench is relatively light on both Zen and M1 compared to some other workloads. They saw consumption in spec
  3. It is impressive. Just noting this is the only benchmark I've seen so far that shows Intel can hold its own single core, and that shows a 4800U as being faster than the M1 in multithreaded loading. Everything else we've seen so far looks even better for the M1 than this does.
  4. Cinebench is probably the most favorable benchmark so far for x86 btw. I'm sure some actual AVX codes will try to be run soon enough which might look silly, but this for now can be mostly seen as a worst case for Apple.
  5. I have a canyon spectral similarly that was around 1700 dollars, but yes. When in the US you walk into Walmart and buy a bike for 100 dollars, and AMD wants to sell the same bike (with probably even less QC work) for 3.5x the price we should call them out on it. (In normal times) you can get ludicrously better and safer bikes for this price point new (ignoring used). Yes upselling isn't uncommon, but I think that isn't an excuse to not call it out. If AMD can't do the work to make the bike even remotely safe (it wasn't safe for road riding even before they brought it into a shop),
  6. I think I agree for almost all of this, I would only caveat that we haven't see yet how Apple will try (if they will try) to target the heavily multithreaded load systems with a higher power chip (the space for example for ARM servers or workstations on Linux now or that a 5900x actually competes in) I want to see a more in depth comparison between the MB Air and Mac Mini because it seems like the doubling of power might not be getting very much performance wise, which in fairness makes the 10W variants even more insanely good. The biggest problem as I see it for AMD i
  7. Yes that is a fair correction now that I've looked back at it. Was looking at the old Zen TR levels which were noticably lower frequency. However my main point was that the majority of the 50W was not core, but IO/interconnects which is literally a design 'feature' of the chiplet set. Even at the 6 or 10W l or 20W even level, it isn't that AMD is more efficient at low thread counts (we haven't seen anything big enough from Apple to care about ring buses or mesh or chiplet), it's just that 50W to 10W is rather non-reflective of relative core strengths. The real issue atm is li
  8. Reminder though that this was a 100 dollar bike being upsold as a 350 dollar bike. Few bad products only bad prices. At the 350 dollar range it's an absolute travesty. All for a (really low quality) sticker.
  9. Thing is, Apple's legal case against Epic literally hinges on there being no exceptions and it being same for everyone, even though we explicitly already knew that was false (Amazon and Netflix already have lower rates). So this will be interesting to say the least.
  10. That is what they seem to be saying, but it's not clear that is what would actually happen. Certainly for laptops, SoCs are almost always more efficient than the same design with flexible arrangement and configurable IO. There are however fundamental limits to the size of a chip that can be made and made reasonably well and AMD literally just has been proving its chiplet design (a move in the exact opposite direction from SoCs) because it dramatically allows them to scale up both individual cores and the total core count. Ofc, AMD does now make some pretty aggressive SoCs in XBX an
  11. I don't actually buy this. Been a while but at the time TR first gen was coming out there were efforts at looking at the per core power consumption and it was maxing around 4-6W. Persuming that stayed similar for Zen 3, that actually puts it very competitive on the literal core end with A14, and it still does have both small ST and massive MT performance leads. The big deal is interconnects. Chiplets is a highly inefficient system for power efficiency of interconnects, and likewise motherboards. But the design goals and targets have been very different. By avoiding SoCs the efficie
  12. It would still be better than the U and H mobile APUs, though not by as much. The real kicker in that source is actually that the Intel chip starts around 50 fps and the laptop cooler is such a piece of shit that it falls down to that mid teens sustained from thermal throttling. On a rough estimate, it looks similar to the 4900HS, which is rather impressive given the power draw difference, though the HS is obviously not optimized the same way, and I do expect given the PS5 and XBSX efficiencies, that AMD could easily match or surpass the performance and efficency on the GPU side at
  13. I am hoping this helps bring real productivity and not shitty games to mobile devices in the next few years, but the lure of microtransactions will probably be too high. Two factors of course... money per user and then Apple has made it as easy as possible if you had even thought of supporting Macs before now.
  14. First, I did say some, not all. Second, doesn't the macbook air completely lack contact on the intel chip? One device is beyond bizzare and completely sandbagging. Third, Intel provides the chips, it is purely the responsibility of the oem to make chassis that handle the parts that are ordered. We wouldn't and don't give this type of excuse to Razer or Dell or anyone else. When it's bad, it's bad. And with the notable exception of the 16 inch MBP, they have been shockingly bad for quite a while. The fact that the 16" MBP maybe runs within 10% clocks of another OEM and w
  15. You are correct, I was specifically talking about the laptops, your point with the desktop chip is also agreed with.
  16. It is impressive, but we really need to remember that Apple has been underdesigning their cooling solutions so badly for so many years now that performance of the same chip in their laptops is way worse than in anyone elses. Almost as if to intentionally make this comparison look stronger.... But change the order a bit, close the gaps, it's still impressive stuff. With or without the "win". (I do think it's extremely important to note that Apple has literally been doing a disservice with how bad most of their chassis designs have been, and literally everyone in the ind
  17. ST comparisons. https://www.anandtech.com/show/16252/mac-mini-apple-m1-tested/4 among others. It does have 2x the Firestorm cores though so yes. Looking right back at it though. I did mess up. Obviously single core power isn't going to be anywhere near the full 6W MT peak for A14 and it was only as high as 15W for M1 at that level as well. Same basic idea though that it's clearly far past the efficiency curve to run at these clocks. (Which tells you something about the mammoth efficiency of mobile phones at low power).
  18. You buy Apple products but want to hit on tesla for being anticonsumer and unethical buisness practices? Anyways. Yes, getting more companies that don't suck as much at these details to make EVs would be great. My understanding is that this problem was being discussed quite a bit over the last few years but NHTSA is pretty much a joke these days. The logging issue isn't something I'd expect on the v3 cars (for a large variety of reasons).
  19. End of the day, I'm not saying these are bad. I'm not saying they are anything short of an incredible effort and demonstration of prowess. Still some lingering questions, but that's true with any new launch. My P1 comment is the best one for my overall thoughts.
  20. "For the rest of the workloads, the A14 generally looks like a relatively linear progression from the A13 in terms of progression, accounting for the clock frequency increase from 2.66GHz to 3GHz. The overall IPC gains for the suite look to be around 5% which is a bit less than Apple’s prior generations, though with a larger than usual clock speed increase." From Anandtech. Clock speed changes along the way have changed things. IPC progression as percentage gains have slowed down significantly.
  21. I didn't say it was worthless. I specifically said it wasn't worthless. I said making general sweeping generalizations in comparative analysis (esp statements that M1/A14 beats X and Y overall) was premature based on just those two tests. And that as a general rule I expect Anandtech to be much more consistent with removing legacy benchmarks as they do for everything else, and if not, explicitly discussing why they didn't (they made a long discussion about this in their last ARM server review, but didnt mention it at all in their A14 discussion and I do find that negligent). Or their A13 revie
  22. Correct, that's what Anandtech found, and I have no reason to doubt them. https://www.anandtech.com/show/16252/mac-mini-apple-m1-tested In the finally more detailed testing analysis.
  23. LAwLz views it as similar results. I don't think I would and leadeater doesn't either. It's 10% difference relative in int workloads, and 5% difference relative in fp loads between the two benchmark sets (how they rank the processors). Given that Apple itself only pushed a 5% ipc improvement this generation, and Intel has commonly come in beneath that gen on gen, it's pretty signficant. Now it wouldn't be a problem in a large testing suite, but it's a bad benchmark to use by itself and make sweeping generalizations about the processors with. Again, Anandtech doesn't do this with literally any
  24. No, he means a scaled up system that can compete with the dGPU workstation boys and a cpu beefed up in cores and similar to compete with Epyc/Xeon. (Or even TR3 / 5950x MT)
×