The value in the ability to upgrade is often misperceived as being worth more than it really is. If you buy something mid-to-top range you are really only going to be able to get a single upgrade cycle out of it out (where there is some substantial marginal benefit for the cost) before the rest of the system starts to obsolete itself. If you watch things closely and the tick-tock cycles are aligned then maybe you get two.
It’s a fun hobby. And if you have the cash to spend on always having the best GPU then you probably don’t care that for $1,000 you only get an extra 10 FPS or so.
These days I can’t even remember how old a machine is let alone what GPU it even has. “Why is this box so slow? It’s four years old? Holy shit, when did that happen?”
Even when I had a rig and was reading Tom’s and Anand daily I don’t think I actually ever upgraded a video card twice. Maybe it happened once? The cycles might have been longer back then, but by the time cards had advanced two generations we were moving from AGP 4x to 8X or to PCI-E or multi core whatever. I could easily just go back in time and buy whatever middle of the road Dell lunchbox was available and just replaced the whole thing every year or two and probably saved money for the same performance.
These days I’ll take a non upgradable whatever it is as long as it has a good warranty.
> The value in the ability to upgrade is often misperceived as being worth more than it really is. If you buy something mid-to-top range you are really only going to be able to get a single upgrade cycle out of it out (where there is some substantial marginal benefit for the cost) before the rest of the system starts to obsolete itself. If you watch things closely and the tick-tock cycles are aligned then maybe you get two.
I don't think that's true in this case.
The upgradeable part would just be an PCIe enclosure. PCIe 2.0 is over ten years old and would still be sufficient for many GPU applications with little performance loss. PCIe 3.0 is eight years old and fast enough that we will likely not see 4.0 for many years in desktops (or mobiles) [also 4.0 needs more expensive materials for all involved planars, so... yeah... not doing that for +-0 %].
It's interesting seeing arguments about how valuable upgrade-ability is, and yet comments like this further outline how upgrade-ability is overrated.
Clearly technology isn't advancing like it used to. I've been running a machine since 2011 when I built it. 15 years ago I maybe would've upgraded every 4 years. Now I'm almost 8 years strong.
Now everyone has different needs, and I think Apple and similar products haven't really been aligned with those needs perfectly, but how much, from now, can graphics REALLY improve, for most use-cases?
It’s never “fast enough.” PCI-E motherboards are actually out now. In three years we will have either PCI-E 4.0 or 5.0 as a standard.
And you’re right—PCIe 3.0 cards out today have plenty of performance for today’s applications. That why you would get a PCIe 3.0 card today. But it doesn’t matter if it’s fixed because you aren’t going to upgrade it.
Actually that's not true. When PCIe 3.0 was introduced, it took quite some time before the (still marginal) performance improvements materialized. We're talking about a few percent at most in games and benchmarks (many are just neutral) between PCIe 2.0 and 3.0, simply because it's not a bottleneck. For not overly data-intensive GPGPU applications (e.g. video things) I reckon that's the case as well. Yes, sure, there are applications you can run on a GPU where PCIe is a bottleneck. PCIe bandwidth benchmarks, for example.
Don't expect PCIe 4.0 consumer-ish (i.e. what this is about) GPUs soon (2019 earliest). It's wholly unclear whether you'll see any performance improvements at all from that — just look at the bus utilization in games and benchmarks, you'll be hard pressed to see more than 10-20 % (which is precisely why running a GPU on a x8 interface (or 2.0) is not an issue).
If it’s not true then you are actually removing the only major benefit of upgradability all together. Just get what’s available today since it’s fast enough and will be for some time.
I was saying that the Blackmagic eGPU in an upgradeable form factor would mean that you get a 200-300 $ (or something like that) GPU enclosure with some electronics in it (Thunderbolt hub, power supply etc.). Then you stick a current GPU into it. This would make sense because GPU replacement cycles are much shorter than the interconnect upgrade cycle. That's why I said that most applications are still fine today with the over ten year old PCIe 2.0.
So just to reiterate: PCIe bus speed is simply not a limiting factor for the overwhelming majority of GPU applications.
It’s a fun hobby. And if you have the cash to spend on always having the best GPU then you probably don’t care that for $1,000 you only get an extra 10 FPS or so.
These days I can’t even remember how old a machine is let alone what GPU it even has. “Why is this box so slow? It’s four years old? Holy shit, when did that happen?”
Even when I had a rig and was reading Tom’s and Anand daily I don’t think I actually ever upgraded a video card twice. Maybe it happened once? The cycles might have been longer back then, but by the time cards had advanced two generations we were moving from AGP 4x to 8X or to PCI-E or multi core whatever. I could easily just go back in time and buy whatever middle of the road Dell lunchbox was available and just replaced the whole thing every year or two and probably saved money for the same performance.
These days I’ll take a non upgradable whatever it is as long as it has a good warranty.