We got some extra budget at my job today. I have the permission to buy a 16 cores 3.1Ghz workstation (HP or Dell). I just wonder about the upcoming Xeon update from Intel… I dont think I can wait but I would like to know what’s coming.
I heard about dual 10 and 15 cores coming in Q3 2013.
thanks!
12-core Ivy Bridge Xeons are coming in September. That’s what the new Mac Pro will have and HP/Dell, etc will also be equipping those. It is a nice upgrade for sure from Sandy Bridge but you’d be fine with dual-socket Sandy Bridge E Xeons like the HP Z820. Don’t buy teh Dell - they are really shabbily made.
I don’t know why you are ragging on the Dell. The workstations from Dell and HP are both fine (wouldn’t be my first choices but they are equivalent). By the way, the Dell Precision T7600 has a 1,000/1,100/1,300 watt power supply which makes me think you haven’t done any homework on this. Last time I checked 1,000+ watts for a power supply is “real” enough for any hardware you could stuff in the case.
I’m sure intel’s new xeon chip will come out by the end of the year, in small quantities and specifically for apple first if we consider apple’s history. It probably won’t be until February or even March until other companies can ramp up production in large quantities for dual-socket systems. I can’t say whether it’ll be worth waiting 7-8 months for, but I don’t think buying a current system would be necessarily bad either.
Bear in mind that even though the the new xeon’s cores are increasing from 8 to 12, there’s a highly likely chance the clock speed per core in multithreading will be lower to compensate for the additional wattage/heat from the added cores. That’s always how it’s been.
The new xeon’s turbo mode will probably boost single-threaded performance by around 9-14%. I think total multithreaded performance will likely increase around 30% real-world and 40% max under certain conditions, but not the 50+% a lot of people might automatically assume going from 8 to 12 cores and to a slightly more advanced architecture. Going from 32 to 22 nm isn’t enough of a jump in thermal and wattage capability to get a 50+% performance gain. Since core count is increasing 50% (which is a lot), clock speed will have to lower to compensate to meet their wattage/thermal constraints for the chips. Going from sandy-bridge to ivy-bridge is a minor performance bump compared to the last xeon’s architecture upgrade going from westmere to sandy-bridge.
Dell and HP’s build quality might not be as good as some, but their systems come with extensive warranties (or can optionally extend) and provide reasonable capabilities. If someone is concerned enough about build quality for a workstation when they open up the machine to add in parts or doing maintenance, they probably should either hand-build their own system since they already have that knowledge of knowing exactly what they want, or pay even more for a more expensive vendor to custom built something that isn’t mass produced on a huge scale.
I didn’t say anything about the PSU. The Dell I was sent was shoddily constructed and the cable and expansion stuff was really gross. I had to remove that from my review since they said it was a preproduction machine but I doubt that changed overnight. Anyway, I reviewed both the T7600 and the HP Z820:
http://arstechnica.com/gadgets/2012/08/as-mac-pro-stagnates-pc-workstations-muscle-ahead/
If I had to buy a PC Xeon prebuilt, I’d probably go with HP. Puget offered me a machine to review which I might use for my Mac Pro 2013 review comparison but I don’t know what their machines are like.
The bit about the power supply was based on earwax69’s comment. Thanks for the review, I always enjoy and learn something new from your articles. To be fair though that wasn’t an apples to apples comparison. The T5600 is a lower class machine in terms of features than the Z820, the T7600 would be a fair comparison. Cost is another story, talking strictly features here (hard drive bays, PCI-Express slots, etc.).
I had to review what Dell sent me in the same price bracket. The T7500 is already more han the Z820 and the T7600 is far more. Last I looked
The 7600 is better and pricier. It come with 32gb of RAM. We would like to buy the RAM by oursleves, The only advantage I see to go with the 7600 is the power supply.
The Z820 really have better customization. They specify that the processor the Dell use need water cooling. (E5-2687w) and also offer the E5-2690 (135w) which Dell dont have.
My boss prefer the Z820 even with the higher cost. We have others z820 and they never had troubles.
How would you benefit overall by having a 16core machine.
We dont have a render farm were I work. They plugged me on a mac pro 12 cores which is nice but dont play so well with Windows. I’ll give the mac pro to my coworker who is till on a Quad and take the Z820. Raytracing will be faster on 16 cores.
On benchmark, the i7 3970x score 12944 while the Z820 get 21647. Not exactly the double but quite a boost on a single machine.
After Effects would be faster on an overclocked i7 but I could not OC at job for sure.
While the budget are here, what would be the best GPU? No quadro. Gaming cards are fine. For C4D, LW and AE CS7.
GTX 780?
The E5-2687w v2 chips will be out soon and will give you 10 core / 20 threads per chip at 3.4 Ghz. Roughly a 45% speed boost over the current E5-2687w. Not sure what the turbo speeds will be, but these will be the fastest of the new crop. The 12 core chips top out at 2.7.
Depends on your time frame. I’d wait 2 months if I could, but the current ones are solid as well.
I’d bet they’ll start trickling out within the next month. They’re being deployed in supercomputers now so they are clearly already being produced in some scale.
While the budget are here, what would be the best GPU? No quadro. Gaming cards are fine. For C4D, LW and AE CS7.GTX 780?
I think the titan would be a better buy than the 780.
I’ll ask the Titan then. perfect.
As for Intel roadmap… 45% more power in 2 months would be worth the wait. I dont think they’ll be able to delay the purchase but I will try none the less. How to know how much to reserve is another thing. Nopbody know the price of the 10 cores xeon.
The new ones will be similar in price. Intel has kept their top end 2-socket Xeons around the same price for several generations.
I’m curious what your evidence is for them being 45% faster since they’re slated to be capable of 225 GFLOPS each compared to the current model’s 185 GFLOPS
http://www.xbitlabs.com/news/cpu/display/20130703223619_Intel_s_Next_Gen_Xeon_Phi_Knights_Landing_to_Deliver_Up_to_3TFLOPS_of_DP_Performance.html
With the E5-2687w v2 being their flagship workstation chip, compared to the current R5-2687w it has 25% more cores, and 7% faster architecture which brings things to roughly 33.75% faster if the chip were to scale perfectly.
Most rendering software doesn’t scale perfectly these days, so I’d think real-world speeds would be more like 30% faster rendering, possibly even less.
E5-2687w
8 cores * 3.1 Ghz = 24.8 ghz
E5-2687w v2
10 cores * 3.4 Ghz = 34 Ghz
So that gets you 38% faster. I added 5-7% estimate for architecture improvements. So 43-45%.
oh yeah, forgot about the 300mhz speed bump
still, LINPACK GFLOPS is about as granular as you can get with fine floating point operations. It’s interesting that intel is rating the chip as 21% faster in a best-case scaling scenario with 225 vs 185.
It makes me wonder if they haven’t been able to increase bandwidth latency as much as would be ideal to keep all those cores fed at those speeds.
Even still, we can’t assume both core count and speed will scale linearly because it never does. It scales worse as they increase, usually around 85-90%. Ivy-bridge architecture isn’t enough to negate the difference. Sandy bridge sure was though when it came out.
My math was a bit off. The 7% estimate would need to be applied to the whole sum since you’d get the arch improvements across all cores, so that’d push it up to 47% if one assumed the same turbo values, and if the numbers posted out there are accurate. Not to mention faster memory interface.
The 225 number seems like a pretty general estimate. That slide could have been produce 2 years ago with as far out as they plan architecture migration.
Still it’s on paper. It will obviously be faster for approximately the same price. Why not wait if you can?
