![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Yep, but if you are looking at it being a platform you want to hold on to for a number of years, it really makes solid case for it self. The extra M.2, the extra lanes, the quad channel memory support are all going to gain usefulness as it ages, and you can always literally double the CPU a few years down the road if more cores are needed and not have to touch anything else. Makes for a very solid foundation for the next few years.
That plus a mid range board would have made a great build and chances are, if I did not already have my Ryzen Builds, that is what I would have gone for on day one. So you don't need the extra lanes, or other benefits of the x399, why do you buy the 1800x over the 1700x or even 1700? That's the problem. The low end side is covered by the 1700, the 1900x then picks it up on the high end.
To an extent yes. But I'm not comparing them in a vaccum. Right now you spend $50 more and you get also get ECC support which is something I really need along with the cores.
The unwashed masses of purchasers buy prebuilt and the 1800x isn't aimed at them. It's aimed at the enthusiast, and there are plenty of other tech sites out there saying the same thing as [H], buy the 1700. Seriously how many tech sites are recommending the 1800x over the 1700x or 1700? It's just not happening. No one is building a box without checking into some of the parts and prices first, especially an AM4 build that requires a focus on memory compatibility. Yes there is going to be handful of people buying the 1800x because they don't want to fuck with the OC, but there is a reason the 1700(x) is outselling it at the major retailers. It's dead to the enthusiast and the mainstream isn't going to pay the price premium or even see it offered. Where is the market for it, really? Rebranded as the 1900x and the gateway to the x399. That's where it still works.
Yeah I'm aware. However, the reduction in cost per core is so low right now that everyone should be upgrading now. When I bough my 6-core Ivy I paid a pretty penny for it (it didn't have ECC either). Definitely more than $500. Probably closer to $650 or more. If I wanted ECC then that meant Xeon and a high clocked six core Xeon was something like $999 and the boards for that were definitely in the $300 range. So it's still a savings if I want a high clocked 8 or more core + ECC. I dont' think you can get ECC on x299.
Yes. Totally, it's the wrong color palette it's still summertime and those are fall colors obviously. I'm assuming you would prefer something in the color blue for the background gradient? I am a Periwinkle man myself. Although some might prefer seafoam as an appropriate color.Does someone find something odd in this slide?
AMD has to do something in anticipation of Covfefelake after all...
Enough that for many applications 1900X will be a quad core.
Proof man, proof. A lot smoke and mirrors on your part with no proof of END results. Latency graphs do NOT speak to end results.
so you have none. Okay. Only graphs have shown latency between parts but not showing end results. Shintai showed a 20%diff between Epyc and 4C intels but nothing showing any software reflecting a 20% diff in results, attributed only to that latency. You see that would be worst case and likely in real world never approach that, realistically prob not even 10% in end results. The lot of you anti-AMDers love to parrot irrelevant data to real world impact. Only those chasing benches would care, however as I and others have stated time-critical software users should go Intel for now as ANY platform in It's infancy is a bad purchase. For the rest of the world, do your homework and decide what best fits your needs and cost expectations.