JustSomeGuy1

About

Banned
Username
JustSomeGuy1
Joined
Visits
60
Last Active
Roles
member
Points
1,172
Badges
1
Posts
330
  • Apple's $5,999 modular Mac Pro now available to order

    The RX580 GPU is old and slow, while the Vega II (Navi W5700X) is at least new, but middle-end at best. Perhaps they are hoping the absurd amount of VRAM compensates for slow compute? Either way, it gets crushed by Nvidia's offerings.

     Combine that with Intel CPUs that have stalled on the same node for five years and get absolutely destroyed by the latest AMD Threadripper CPU and you now have the most expensive obsolete on arrival workstation on the market. Did someone get the hardware decisions exactly backwards over at Apple?

    Why release literally the worst of each possible hardware camp in a single product? AMD CPUs and Nvidia GPUs would have made this MacPro a beast.
    I've already made the point that AMD poses a big problem for them down the line, but the truth is that Intel still owns the workstation market. Not for long, I expect, but at this point it's still the safe choice being made by most people putting down 20-60k per workstation.

    OTOH, you don't know what you're talking about re: the Vega II. It's *not* Navi. And how it performs is *the* big open question. Well, that, and how much/how fast will vendors optimize their code for it? The jury's out. And yes, I'd feel happier if Apple supported NVidia again. But this story's not written yet. Let's see what happens.

    watto_cobra
  • Apple's $5,999 modular Mac Pro now available to order

    The pricing is very interesting, and much cheaper than I'd expected, when you look close.

    First of all, CPU pricing is *surprising* - the cost of the chip upgrade from 8 to 28 cores is only about $300 higher than the difference in Intel's list prices. Obviously Apple is paying less than that, but still, this is a remarkably low mark-up! Put another way, Apple's margin on the nnMP configured with 28 cores is much lower than their margin on the 8-core unit. And I don't expect many purchasers to go with the 8-core unit. So, if they're Ok with that lower margin, why price the base at $6k? It's a bit of a mystery.

    Secondly, SSD prices are reasonable, assuming you're willing to say that the Apple SSD performs on a par with top-grade SSDs like the Samsung 970 Pro. Those list at $300/TB, the same per-TB cost as upgrading from 2 to 4TB. Of course you can get quite good performers- maybe marginally better than Apple's but without the T2- for $150-$200/TB. Nothing stops you from buying one (or four) of those, though, sticking them on a $100 PCIe card, and dropping them into your nnMP, if you want to. All in all, Apple SSDs aren't cheap, but the added expense is way more reasonable than it was on, say, the last 15" MBP.

    Memory is ridiculous. But that's easy to do yourself. I would expect that many users will order memory separately - even large corporate users. Those that don't are going to get reamed, but (surprise!) by a bit less than they would be if they were buying from Dell or HP.

    The video cards are a big question mark. We won't know what they're worth until we start seeing a bunch of benchmarks. Apple's valuing the 580 at $400 which is high but in line with other name-brand vendors. Of course the interesting cards are all much pricier.

    In short, if you're willing to buy your own memory, you can get a seriously roomy nnMP for $19k: 28 cores, 4TB SSD, one dual-GPU "Pro Vega II Duo", and the default 32GB RAM. Add $900 for 6x32 (192GB). Dump the 4x8GB, or buy another two for $100 and use all six for an extra 48GB. Now you're at $20k, and your Mac has 240GB RAM. Double your GPU for another $5k, if you like.

    Overall this seems more than reasonable for users who actually need this power. If we lived in an Intel-only world, it'd be a smash hit at an excellent price.

    The one problem is AMD. They're just starting to make inroads now but in six months they'll be everywhere. And I expect Dell and HP will be selling them in their workstation product line (whereas now Dell seems to be selling them only as "servers" and I can't find them at HP at all). The 64-core EPYC 7702 crushes the Intel 28-core chip, often by a factor of 2, while costing half as much. And you can get dual-chip systems if you really want a *lot* of cores.

    This isn't going to make the nnMP obsolete right away - after all, Intel will still be inside a large majority of all systems, and in that class, the nnMP is a winner. But as a halo product, it won't be "the most powerful" anything. It'll be "half as powerful" or "a quarter as powerful" as top AMD single- and dual-chip workstations. That's not so great.

    What Apple does next will really be more interesting than this nnMP. Will they build a 2020 model? Whether AMD or Intel, will they offer something better? Or do they let this thing sit there and shrivel in the harsh light of rapid tech advances, as they did with the nMP? If the former, I think they've got a great future on the high end. And if the latter, then they're done in that market, for at least the next 10 years, and probably a lot longer. Nobody will ever trust them again.

    Time will tell.
    HereForTheHardwarejdb8167watto_cobra
  • How Steve Jobs saved Apple with the online Apple Store

    R Lubin said:
    Macintosh clone company, Power Computing sold their poorly made but inexpensive Mac clones solely online[...]
    I bought about a dozen of them for my company. They were great- cheaper and better than Apple's Macs at the time. So much better than I "stole" one to use in my home office and replaced it with my personal genuine Apple Mac. It was surely the right move on Jobs' part, but at the time I was seriously aggravated when he shut down licensing.
    watto_cobracornchipBart YFileMakerFeller
  • Apple partnering with tech giants for big clean energy buy

    toysandme said:
    How many more millions of birds are these companies willing to kill and remove from our environment?
    Nowhere as many as cats kill.
    When a cat does it, it’s nature’s way. When a windmill farm does it, it’s an abbatoir.
    Are you an uneducated bird lover, or are you a tool of the carbon lobby?

    Climate change will drive entire species to extinction (never mind what it'll do to us). The bird deaths due to windmills are unfortunate, but a much lesser price to pay. No person (or indeed any creature) can live entirely without footprints; unless you're willing to self-euthanize, you're going to have impacts on other entities, birds or others. Are you a vegetarian? Are you willing to attempt to legislate that everyone else become one? If not, your protests are a transparent deception.

    All that said, I'm strongly in favor of developing next-gen nuclear. It's good to have options. But there is no possible way we're going to have any significant portion of our energy use served by new nuclear reactors in the next 10 years, even if we start construction right now. You want nuclear? Good, go for it, but in the meantime we need other options. And in any case, it is not a sensible investment for Apple or other companies. It's speculative and very long-term, whereas solar and wind can be put in place in a year, give or take.
    auxiominicoffeewatto_cobra
  • Editorial: Samsung's new Exynos 990 isn't fast enough to maintain its custom M core

    auxio said:
    Rayz2016 said:
    The next article will take a closer look at how Apple was able to sneak past larger more entrenched silicon experts at Intel, Qualcomm, and Samsung while the media doubted its progress and bet that rivals would easily catch up.

    Yup, I'm a lot more interested in how Apple did it, rather than how Samsung couldn't. But I guess it's a question of context.

    My guess is that it's due to a couple of factors:

    • The market for mobile devices is far larger than PCs (Intel's core market), which gave Apple a lot more money to work with.  Apple bet big on mobile and it paid off
    • They didn't have to compete with Intel head-on like they did in the PowerPC days.  Mobile devices weren't expected to do everything PCs do.  So they could start from scratch and scale up the performance over time.  Unlike Intel who were still entrenched in PC CPU design (and trying to make PC CPUs more power efficient for mobile)
    • They control the whole stack: from CPU to OS to app APIs.  Even if Intel came up with a great new mobile CPU design, they still have to work with mobile OS creators to ensure they're making use of the new features.  Same with Qualcomm.  Samsung was the only one who could compete here, but they still have to work with Google to have Android take advantage of their CPU features
    The first couple of factors are clearly a big part of this. Your third, however, is off-base. It's correct, and important in other contexts, but not really very relevant here. It's not like Apple's huge lead in core performance is due to their ability to win big on specialized applications that depend on holistic product design. They kick everyone's ass on basic benchmarks (SPEC2006, for example) that measure very fundamental aspects of a CPU's performance. APIs aren't relevant, and even the OS takes a back seat (or should).

    At the level of those benchmarks, the OS shouldn't really be a factor, having a real impact only if there's a big screwup. For example, last year's (or was it two years ago's?) Galaxy with Exynos was severely throttled by bad DVFS, in a way that the QC variant wasn't. This had a significant impact on some benchmarks. But other than that, the OS shouldn't have much impact on low-level performance measurements.

    Every time someone trots out the "whole stack" argument, they're missing a key point: Apple's silicon is just *that much better*. At this point, the only performance competition Apple's got are the high-end i7 and i9s from Intel, and the Zen2 Ryzen 3xxxs from AMD (I'm ignoring small-market CPU's like POWER). Neither of those, of course, comes even close in power efficiency.
    chaickanetmagewatto_cobra