Bolt II

I'm going to guess that the Strix isn't really available for the Bolt II and really there because other computer configurations have it.

But if it is, they would have needed to have redesigned the GPU bracket, and probably removed the top of the chassis to get the GPU in with the PCI power connectors on. Also, the PCI power connectors will protrude more on the Classified board than on the Strix, as the PCB on the Classified extends the full height of the shroud, whereas the Strix's does not and the are recessed some relative to the height of the shroud, giving it a little be extra space.

If the Strix is truely available and you want to best-of-the-best type VGA, you may want to consider the ASUS Matrix Platinum, as it has a very similar set-up to the Classified with PCI power connectors recessed a bit. I'm not sure if the fans on the Matrix Platinum can spool down to zero, though, like the ACX 2.0 fans can...

DS employees have mentioned in other threads that keeping an exact configuration for each model is too much work and they'll put up a few common configurations and review each configuration as they come in to make sure they work. I'm not sure I completely understand, technically, but logically I agree that configuration control is a pain...if they correctly set up a database for the options, they should be able to tick which options should come up for each chassis (just have each chassis type in the column and all of the options for the rows; reset the form if the chassis type is changed).

It is interesting that they've increased the height of the chassis...maybe they received enough interest in VGA cards like the 980 Classified and STRIX to warrant a change like that. I did however notice they completely removed the ODD slot (It looked like they just photoshopped out the ODD face plate from some of the pics on the Bolt II page), which is a pity...I think they should have gone with an integrated slot and moved to a slot-load DVD player. So I guess I'm happy I got mine when I did; I'll have a smaller machine, which is what I want from a SFF chassis.

http://www.digitalstormonline.com/forums/bolt-ii-slot-loading-drive-tidf23983/

Confirmation that the slot for the ODD has been removed, as well as the height of the chassis has been increased to accommodate taller graphics cards.

The R9 300 series will be announced in the near future.  It should have High Bandwidth Memory (HBM) and give a serious challenge to the Green Team (even their Ti models).  The Ti models usually trail the initial cards by 6 months or so, so they may be released around Q2 next year; so they're a few months out, still.

But there will always be a 'next' thing to come out...you have to pull the trigger at some point.  HBM will be a fairly big step up for graphics...the PCI-e interface will be a limiting factor once again, unlike the current gen, which can barely saturate a 3.0 x4 connection.  Other advancements are no where near as big, in terms of performance gains.  DDR4 has really only shown to be better in low power and short bursts, but the price of it is still quite high, and DDR3 performance can still equal it for all but the most extreme edge cases.  Broadwell-K will be coming out sometime next year.  Those processors should still be compatible with the LGA-1150 sockets and 97-series chipsets.  The low power Skylake chips are also supposed to be introduced, as well, but they won't match the performance of the Broadwell chips; just have additional microarchitecture tweaks, new ISA support, DDR4 support, and whatever else Chipzilla decides they need to do to compete with the high-end ARM market.


Edited by  - 26 Dec 2014 at 2:55pm

Maybe there should be a name change, Bolt11.5 or Bolt111

Originally posted by Corvosin Good points. From what I've seen a 980 oc'd can achieve 60fps at 1440p in most games. I may just go for EVGA's GTX 980 SC ACX.

FYI, EVGA came out with a higher binned GTX 980 FTW recently...don't know what the availability for it is like, though.

The CES 2015 event is coming up soon. We'll see if there are any more goodies to be released. If not I'll pull the trigger next week.

Looks like the new Aventum III...   

What is this "background" of which you speak?  All I see is a blue, red and black goddess suspended over a sea of awesomesauce. it's white acrylic tubing calling to me with a siren's song to dash among the rocks of want.

Well I think it's safe to say Nvidia will not be releasing a better card anytime soon. I think EVGA's GTX 980 FTW is the best choice + it's available. Thinking of purchasing my Bolt II this weekend. I'd just like a confirmation that DS will email me about special ordering the GPU and not just put a default 980 in there.

Originally posted by  http://www.digitalstormonline.com/forums/bolt-ii-slot-loading-drive-tidf23983/ The R9 300 series will be announced in the near future.  It should have High Bandwidth Memory (HBM) and give a serious challenge to the Green Team (even their Ti models).  The Ti models usually trail the initial cards by 6 months or so, so they may be released around Q2 next year; so they're a few months out, still. But there will always be a 'next' thing to come out...you have to pull the trigger at some point.  HBM will be a fairly big step up for graphics...the PCI-e interface will be a limiting factor once again, unlike the current gen, which can barely saturate a 3.0 x4 connection.

we should see AMD's 380x and 390x fairly soon (1st half of 2015), you never know with AMD, but i think it's worth waiting for (I am).  AMD should have a full year on Nvidia for their tardiness in adopting HBM.  we'll see Nvidia try to counter AMD with the Titan II (or whatever they're going to call it) and 980ti, but they'll both be on (inferior) GDDR5 memory.  

Originally posted by  HBM will most likely still be GDDR5, just on a stacked die with the graphics processor...they are both manufactured using CMOS processes (AMD may license design IP from one of the major NAND vendors and incorporate it into theirs). The benefit of putting the memory on die is that you can have a huge (very wide, in industry parlance) interface with the memory, as well as much shorter traces with known and tightly controlled signalling that you can really push on. But yes, NVIDIA will probably be playing catch-up up until Pascal is released next year. Another question, though, with HBM: will the PCI-e 3.0 interface become completely saturated? But with Graphics cards able to work well with x8 or sometimes even x4 connections, there should still be a substantial performance boost, regardless.

i hesitate ever using the x4 connection for discrete graphics, but regardless it will take a new mobo design, utilizing a faster interconnect bus like Nvidia's NVLink, to fully feed this new architecture, something desk top users won't see for quite a while.  i suspect we will see very nice gains in energy delivery and efficiency while still using the pci-e 3 and even 4 bus.

one thing i'm concerned about is if XDMA is enabled on the new cards, how will that effect bandwidth contention across the pcie bus?  right now it's not a problem because they can account for any latency by planning frame transfers around it,  and current generation cards don't saturate the x16/x8 lanes anyway. 

AMD's hypertransport is similar to QPI and NVLink. 

the way i understood it desk top cpus/gpus (Pascal) will connect over a dual PCIe/NVLink and the server community will have NVLink exclusively.  this should provide both cpu-gpu and gpu-gpu connectivity. but you're right, servers (IBM) will see the first use of this tech (i think they've actually partnered up with Nvidia to use NVLink on their next power chip).    

it's kind of ironic that Nvidia will be relying on AMD's HBM tech to take advantage of their superfast serial interconnect tech.  i just wonder if AMD will license NVLink from Nvidia or have something similar in the pipeline of their HSA/GCN. 

No problem. I posted my configuration and thoughts on noise in the second to last paragraph in my review. Another guy also moved the fan control over to the motherboard as well.

I've spoken with John and I'm able special order the card I desire. Unfortunately I cannot return the system due to the special order. I have not intention to return the system but does that include defects?

Originally posted by Corvosin I've spoken with John and I'm able special order the card I desire. Unfortunately I cannot return the system due to the special order. I have not intention to return the system but does that include defects?

I believe that means that you cannot return it for a refund. However, I would think that this in no way affects the warranty for parts or labor. I can attest that they will work with you to resolve any issue you may have with your machine...I have definitely appreciated it and has made my ownership enjoyable.

Originally posted by Alex Basically if we special order a card, if it goes defective, you will have to work with the vendor to get it serviced, we won't be able to offer you a credit on it. In addition, let's say you purchased a system from us and you ordered 3x GTX 980's, if you end up having hardware issues that we have confirmed to be defective and you were getting a refund, we would not be able to refund you on the cards, you would receive them back from us.

Makes sense. So basically anything involving the card I special order you cannot cover? Like ordering the system with no card essentially. You order and install the card and the rest is up to me?

I'm quite set on EVGA's GTX 980 FTW ACX. Do you special order these cards from the manufacturer (EVGA in this case)?

Alex, I believe he was asking if you procure the card directly from EVGA or go through a middle man like Newegg Business or similar.