New Posts  All Forums:

Posts by KyadCK

I personally got a WS-C4948-E, though all ports including the SFPs are 1gbps only. In the end I swapped it out with an IBM G8000F with a dual SFP+ addon and a G8124, as I expanded my backbone to 20gbps. Note that none of these are PoE. Please don't assume these are the best options for you, as there may be others that are more cost efficient for you. The easiest way to know what a switch can do is to just google search for the model and look for the product sheet, which...
Yes, though the SFPs should be 1gbps and it should be a PoE switch.
Some subs from @Hoodcom. TV is 55" and the "little" speakers are CV E-312s, which have 12" woofers. Their job is actually just mid-bass for the theater, though the 18" cones look like they could kill small critters.
Yes... You do...If you are found to be using someone else's patented design that you did not licence, you will be sued both for an extreme amount of money, a portion of money from all infringing products sold, a ban on selling those products in whatever countries you are suing them in, AND possibly forced to sign a licencing deal.If they want to not be sued, Intel needs to either licence GPU IP (like they did from nVidia), buy the GPU from someone who does have the...
NVMe is a storage protocol for PCI-e, nothing more, same as AHCI was for SATA (which also supported ATA mode) when it was introduced. NVMe is not restricted to M.2 slots, and also works over U.2, standard PCI-e slots, and even SATA Express.M.2 on the other hand is not just an SSD standard, but a micro PCI-e standard, and even SATA standard as a whole. It supports WiFi cards, WAN cards, Bluetooth cards, SATA AHCI SSDs (mSATA), PCI-e AHCI SSDs, and PCI-e NVMe SSDs. It is a...
And then get sued for patent infringement.You don't licence IP to learn how it works, you licence IP to be allowed to use the design.
https://en.wikipedia.org/wiki/NVM_ExpressNo, but only because you are combining terminology. NVMe is to M.2 as AHCI is to SATA.It is possible we could make a low power M.2 2280 or U.2 GPU, but M.2 slots do not natively support any form of cooling, and even SSDs that peak at 15W will thermal throttle. The GPU would be outclassed by even Intel's weakest iGPUs though, making it an awful waste of 4 PCI-e lanes. Also no display outs, so it's just be a very weak compute card.
nVidia won't share. With the exception of the Coffee Lake/Vega SoC, Intel can not use Freesync, as it is an AMD hardware and driver suite combination.However, Intel has already said they will support AdaptiveSync, which potentially makes it compatible with Freesync certified monitors.
The far more important part is it has an HBM stack.2-4GB of 128-256GB/s means that this is a REAL GPU, not choked out by a lack of bandwidth since DDR4-2400 is under 40GB/s. This could kill nVidia's low end chips.Na, the power will have to be kept low so I'm expecting 128GB/s out of the HBM and as a result I'm expecting a 1280-shader part or so clocked nice and low for the efficiency. Otherwise they'd just go dGPU as usual with a 1060 or something.7870 in an ultrabook with...
New Posts  All Forums: