Slashdot Mirror

I've noticed this as well.

My HTPC struggles with 1080p encoded x.265 video, so I don't believe there is any real benefit to the consumers if we need to upgrade our equipment.

For the record, my HTPC has an AMD Athlon 5370 APU.

Intel makes a number of ultra efficient x86 cpu's now.. I own several... The $49.99 Winbook TW-700 Windows 8.1(not RT) tablet with full size USB, micro hdmi, seperate charging port consumes less than 2 watts when running idle with screen off; screen on 3-4 watts. Motorola Droid2 performance playing the same videos/full load uses very similar(screen not withstanding)

https://www.cpubenchmark.net/p...

The problem today with Microsoft; controlling 90% of the market is they are rounding the wagons and trying to build a walled garden to a locked down proprietary system(metro, and single screen android like apps). It's all part of uefi, secureboot and lack of available information to create drivers for alternative OS's.

Unified system could work quite well, problem is everyone is attempting to make one for the phone first, then moving it to the desktop. I've used full desktop window managers xterm/vnc'd in on phones just fine.. Industry is backwards it seems. Win32/desktop applications are perfectly usable with pinch-zoom or slight display modifications.

I'm someone who would love to buy a modern day OQO. Imagine a Droid 3(photon q) case(slide out keyboard, hdmi, usb(throw on 2nd), removable battery, mircosd, sim, headphone jack) but with M-5Y10(or something similar/more power efficient) with legacy bios that could still run everything... Phone functions could be run through a program on a regular Windows XP/7/8.1 desktop.

My trusty i7 920 @ 3.8 Ghz was pretty much the plateau for Intel chips. Sure I could 'upgrade' to a 4.5 ghz for $1800 but really no reason to.

You're way out of touch with modern CPU performance. Modern CPUs have better performance per clock cycle, thanks in part to things like newer SSE2 instructions. Anyway, I've filtered https://cpubenchmark.net/common_cpus.html a bit for you:

• Intel Core i7-5820K @ 3.30GHz 12,987 $389.99 (33.3 passmark/$)
• Intel Core i7-4790K @ 4.00GHz 11,206 $339.99 (32.9 passmark/$)
• AMD FX-9590 Eight-Core 10,261 $224.95 (45.6 passmark/$)
• Intel Core i7-4790 @ 3.60GHz 10,019 $305.00 (32.8 passmark/$)
• AMD FX-8350 Eight-Core 8,945 $149.99 (59.6 passmark/$)
• Intel Core i7-920 @ 2.67GHz 4,994

I've included your CPU for performance comparison. It's your CPU only has half of the performance of a modern CPU that only costs $225 - $305. Also note that your i7-920 only supports up to 24GB of DDR3, so upgrading to a new CPU will also mean doubling your RAM speed.

A quick price search shows that you could double your machine's performance for about for $425 on NewEgg (FX-9590 + ASRock 990FX Extreme9 + 32GB DDR4). It should only cost you about $80 more to stick with Intel. You can probably reuse your old case / keyboard / monitor / mouse, and if your GPU won't fit in that motherboard, then you really need a new one anyway.

p.s. The FX-9590 is only 48% faster than your CPU in single-threaded tests, but the i7-4790 is 97% faster than yours in single-threaded.

I just bought a Dell M6700 for cheap. Pulled the CPU and replaced it with a Intel(R) Core(TM) i7-3940XM CPU @ 3.00GHz. Replaced the monitor with a 1080p one for under $100.

For under $1000 I have a 17" laptop with competitive benchmarks, room for 4x internal hard drives (3x 2.5" 1x M.2) and 32 GB of RAM. It runs FreeBSD and Linux Mint just fine. Firewire, USB3, Display port, VGA, HDMI and serial/parallel ports when connected to a dock.

It's all I could ask for in a mobile workstation.

https://www.cpubenchmark.net/
https://www.spec.org/cpu2006/r...

Are you sure about that? Everything I have read has stated that, while clock rates have plateaued, CPU performance increases pretty well every revision. Just because the new CPU has the same clock rate, doesn't mean that it is the same speed. I recently had to compare a HP Gen 7 server to Gen 8, only 2 years or so apart, same clock rates on the Xeons, but the newer one had literally twice the performance of the older system. Same number of cores, same clock rate, but the logic improvements doubled the performance for workloads.

This only mattered in our usage as these were for busy Exchange Hub servers, where they are CPU bound by the virus scan engine, so a double in performance is a double in the mail queue it can handle.

It's pretty telling when the CPU single-thread desktop performance leader, the i7-4790K, is almost two years old. That used to be an eternity in silicon fab. Intel is busy on the server side cramming ever-more cores into their Xeons for high-density server rooms and reducing power consumption on the mobile side. The market (and Intel, who in part sets the market) has decided the Devil's Canyon is apparently fast enough for any single-threaded work you'll ever do. That doesn't help those of us who count on software whose vendors haven't yet implemented / optimized multi-core support in their apps.

AMD has been missing out on these benchmarks for a long time: http://www.cpubenchmark.net/hi...

Slightly better CPU than my old laptop.

And the Kickstarter page says it can run Linux, Android or Windows.

I might buy one and install XP on it.

intel's illegal activities is not why amd has fallen over the last decade... intel hasn't looked back since core architecture first was released. up until that point, amd was cruising and taking the market with superior processors.... but core changed *everything* -- and is still the reason amd sucks ass today. they still can't beat it as intel continues to refine the architecture and shrink its manufacturing processes.

http://cpubenchmark.net/market...
this is based on passmark's user base, not sales, but it follows market trends pretty well.. guess when core was released

Clock for clock the Pentium III was a better CPU than the Athlon XP.

You are on crack. A Pentium 4 can't get anywhere close to the performance of ANY Core 2 or above CPU.

Intel Pentium 4 3.80GHz
Score: 493

Intel Core2 Duo E4300 @ 1.80GHz
Score: 1054

Intel Core i7-5960X @ 3.00GHz
Score: 15,997

I can't believe there are still people who think MHz is an indicator of performance.

You are on crack. A Pentium 4 can't get anywhere close to the performance of ANY Core 2 or above CPU.

Intel Pentium 4 3.80GHz
Score: 493

Intel Core2 Duo E4300 @ 1.80GHz
Score: 1054

Intel Core i7-5960X @ 3.00GHz
Score: 15,997

I can't believe there are still people who think MHz is an indicator of performance.

You are on crack. A Pentium 4 can't get anywhere close to the performance of ANY Core 2 or above CPU.

Intel Pentium 4 3.80GHz
Score: 493

Intel Core2 Duo E4300 @ 1.80GHz
Score: 1054

Intel Core i7-5960X @ 3.00GHz
Score: 15,997

I can't believe there are still people who think MHz is an indicator of performance.

You're in denial and you don't know what you're talking about. Core count and clockspeed don't equate to performance across different models and brands of CPUs. Here are a few Core i3 CPUs that are faster and much lower power than your FX 4350.

Intel Core i3-6300
Score: 5894
TDP: 51W

Intel Core i3-6320
Score: 5773
TDP: 51W

Intel Core i3-4370
Score: 5545
TDP: 54W

Intel Core i3-6100
Score: 5494
TDP: 51W

Intel Core i3-4360
Score: 5493
TDP: 54W

Intel Core i3-435
Score: 5264
TDP: 54W

AMD FX 4350 -Your CPU
Score: 5198
TDP: 125W

You can buy an MSI LGA 1150 motherboard for $40 and a Gigabyte LGA 1151 motherboard for $56 on Amazon right now.

You're in denial and you don't know what you're talking about. Core count and clockspeed don't equate to performance across different models and brands of CPUs. Here are a few Core i3 CPUs that are faster and much lower power than your FX 4350.

Intel Core i3-6300
Score: 5894
TDP: 51W

Intel Core i3-6320
Score: 5773
TDP: 51W

Intel Core i3-4370
Score: 5545
TDP: 54W

Intel Core i3-6100
Score: 5494
TDP: 51W

Intel Core i3-4360
Score: 5493
TDP: 54W

Intel Core i3-435
Score: 5264
TDP: 54W

AMD FX 4350 -Your CPU
Score: 5198
TDP: 125W

You can buy an MSI LGA 1150 motherboard for $40 and a Gigabyte LGA 1151 motherboard for $56 on Amazon right now.

You're in denial and you don't know what you're talking about. Core count and clockspeed don't equate to performance across different models and brands of CPUs. Here are a few Core i3 CPUs that are faster and much lower power than your FX 4350.

Intel Core i3-6300
Score: 5894
TDP: 51W

Intel Core i3-6320
Score: 5773
TDP: 51W

Intel Core i3-4370
Score: 5545
TDP: 54W

Intel Core i3-6100
Score: 5494
TDP: 51W

Intel Core i3-4360
Score: 5493
TDP: 54W

Intel Core i3-435
Score: 5264
TDP: 54W

AMD FX 4350 -Your CPU
Score: 5198
TDP: 125W

You can buy an MSI LGA 1150 motherboard for $40 and a Gigabyte LGA 1151 motherboard for $56 on Amazon right now.

You're in denial and you don't know what you're talking about. Core count and clockspeed don't equate to performance across different models and brands of CPUs. Here are a few Core i3 CPUs that are faster and much lower power than your FX 4350.

Intel Core i3-6300
Score: 5894
TDP: 51W

Intel Core i3-6320
Score: 5773
TDP: 51W

Intel Core i3-4370
Score: 5545
TDP: 54W

Intel Core i3-6100
Score: 5494
TDP: 51W

Intel Core i3-4360
Score: 5493
TDP: 54W

Intel Core i3-435
Score: 5264
TDP: 54W

AMD FX 4350 -Your CPU
Score: 5198
TDP: 125W

You can buy an MSI LGA 1150 motherboard for $40 and a Gigabyte LGA 1151 motherboard for $56 on Amazon right now.

You're in denial and you don't know what you're talking about. Core count and clockspeed don't equate to performance across different models and brands of CPUs. Here are a few Core i3 CPUs that are faster and much lower power than your FX 4350.

Intel Core i3-6300
Score: 5894
TDP: 51W

Intel Core i3-6320
Score: 5773
TDP: 51W

Intel Core i3-4370
Score: 5545
TDP: 54W

Intel Core i3-6100
Score: 5494
TDP: 51W

Intel Core i3-4360
Score: 5493
TDP: 54W

Intel Core i3-435
Score: 5264
TDP: 54W

AMD FX 4350 -Your CPU
Score: 5198
TDP: 125W

You can buy an MSI LGA 1150 motherboard for $40 and a Gigabyte LGA 1151 motherboard for $56 on Amazon right now.

You're in denial and you don't know what you're talking about. Core count and clockspeed don't equate to performance across different models and brands of CPUs. Here are a few Core i3 CPUs that are faster and much lower power than your FX 4350.

Intel Core i3-6300
Score: 5894
TDP: 51W

Intel Core i3-6320
Score: 5773
TDP: 51W

Intel Core i3-4370
Score: 5545
TDP: 54W

Intel Core i3-6100
Score: 5494
TDP: 51W

Intel Core i3-4360
Score: 5493
TDP: 54W

Intel Core i3-435
Score: 5264
TDP: 54W

AMD FX 4350 -Your CPU
Score: 5198
TDP: 125W

You can buy an MSI LGA 1150 motherboard for $40 and a Gigabyte LGA 1151 motherboard for $56 on Amazon right now.

You're in denial and you don't know what you're talking about. Core count and clockspeed don't equate to performance across different models and brands of CPUs. Here are a few Core i3 CPUs that are faster and much lower power than your FX 4350.

Intel Core i3-6300
Score: 5894
TDP: 51W

Intel Core i3-6320
Score: 5773
TDP: 51W

Intel Core i3-4370
Score: 5545
TDP: 54W

Intel Core i3-6100
Score: 5494
TDP: 51W

Intel Core i3-4360
Score: 5493
TDP: 54W

Intel Core i3-435
Score: 5264
TDP: 54W

AMD FX 4350 -Your CPU
Score: 5198
TDP: 125W

You can buy an MSI LGA 1150 motherboard for $40 and a Gigabyte LGA 1151 motherboard for $56 on Amazon right now.

As I understand it, an Atom CPU roughly matches the performance of a Pentium 4 CPU clock for clock. This means an old application that runs well on a P4 will also run well on an Atom.

That is demonstrably wrong. Using the benchmark, Passmark, a 3GHz Pentium 4 has a CPU Mark of 358 while the previous -generation 1.6GHz Atom Z3795 scored 1684. The comparison site didn't have the Passmark of this newer Atom CPU, but the CPUs are similar enough.

As I understand it, an Atom CPU roughly matches the performance of a Pentium 4 CPU clock for clock. This means an old application that runs well on a P4 will also run well on an Atom.

That is demonstrably wrong. Using the benchmark, Passmark, a 3GHz Pentium 4 has a CPU Mark of 358 while the previous -generation 1.6GHz Atom Z3795 scored 1684. The comparison site didn't have the Passmark of this newer Atom CPU, but the CPUs are similar enough.

Interesting.

I went and added my i7 from 2012 just of curiosity too

It seems you are mistaken: http://www.cpubenchmark.net/co...

My ancient Core i7 2700 from the end of 2011 is still considerably quicker.

according to this, the new processor would be slightly slower than your old one. plus, your old one can be easily overclocked.
http://cpubenchmark.net/compar...

Pretty slow.

"(since games haven't been CPU bound in years)"

You obviously haven't played any system stressing games... Most games are not multiuthreaded, so do not benefit from AMD's main competitive edge. Not to mention AMD chips run hotter and use more power than a comparable intel. Never a good sign of good design .

Then you imply that intel rigged ALL the benchmarks they are in because there is a conspiracy and the US DOJ should get involved....

Like this right, this is rigged by intel?
http://www.cpubenchmark.net/hi...
(The first AMD cpu ranks in at 59th... with double the cores of the nearest i7)

The pendulum switches back and forth all the time, no point in being a fanboy for either franchise. Intel has been dominating since core 2 duo / core 2 quad and they continue to do so. I had dual opterons before that. Brand loyal is stupid, benchmarks cant all lie.

Just go to cpubenchmark.net.

You might want to check CPUMark scores for those...

AMD Athlon 64 X2 Dual Core 4000+
Intel Core i7-5557U

You might want to check CPUMark scores for those...

AMD Athlon 64 X2 Dual Core 4000+
Intel Core i7-5557U

Intel is slapping i5 (and i7) on some pretty slow chips these days...

Z3735G (which I assume is about the same if not better as the Z3735F)

Yep, that's correct, G is a nudge faster than F.

CPUMark: Z3735F, Z3735G.

Z3735G (which I assume is about the same if not better as the Z3735F)

Yep, that's correct, G is a nudge faster than F.

CPUMark: Z3735F, Z3735G.

In other news, AMD stock goes through the roof.

You're acting like China won't still be able to get their hands on a stack of Xeons any time they want to with Lenovo and Foxconn both sitting inside their borders. Plus, AMD can't deliver anything close to Xeon performance, much less at the same power rating. Nobody wants to dump 10MW into a computer room and then evacuate that heat if they can do the same job in 6MW with 2.5x the performance*.

Looking at this really should shed some light on where high-end computing sits right now - AMD isn't even in the top 50, and anyone building nuclear weapons number crunchers doesn't give a damn about price.

* Intel Xeon E5-2699V3 averages 24601 on CPUMark in 145 watts TDW, where AMD FX-9590 8-Core scores 10273 on the same benchmark, in 220 watts.

but it doesnt have to be like this, its because the manufacturers have been pushing the ultrabook class cpus into everything because they are cheap.

for example, you used to be able to get this processor in a 600-700$ laptop
http://www.cpubenchmark.net/cp...

That is 3x the performance of these U series processors. It even matches my 2600k. However since the U's are much cheaper, and as stated in the summary, most users wont even know the difference since all they do is surf and watch movies

So, I personally don't follow performance numbers too much these days, but I just went and did a comparison of this "new" system against my current desktop (most components are 4-5 years old inside)

Theirs:
http://www.cpubenchmark.net/cp...
http://www.videocardbenchmark....

Mine:
http://www.cpubenchmark.net/cp...
http://www.videocardbenchmark....

So, the thing barely tops my "ancient" (by today's standards) desktop computer for CPU performance. It has half the RAM (even my old 10" netbook has 8GB DDR3)

Really, I think I'll just label this article as another #Slashvertisement.

So, I personally don't follow performance numbers too much these days, but I just went and did a comparison of this "new" system against my current desktop (most components are 4-5 years old inside)

Theirs:
http://www.cpubenchmark.net/cp...
http://www.videocardbenchmark....

Mine:
http://www.cpubenchmark.net/cp...
http://www.videocardbenchmark....

So, the thing barely tops my "ancient" (by today's standards) desktop computer for CPU performance. It has half the RAM (even my old 10" netbook has 8GB DDR3)

Really, I think I'll just label this article as another #Slashvertisement.

It's amazing just how fast it runs on a five-year-old i7 950 PC with 16GB of RAM.

Yeah, it's five year old, but even today a very fast CPU. Has a CPUMark of 5600. High TDP of course.

"double the cores" does not make a computer 10x faster.

Sure it does, in binary.
But in all seriousness, that is the point I am making, I had 2 times the cores, the same clockspeed, but independent reviews spec the new processor as 10X more operations than the old one. Obviously we have made significant advances in transistor density, instruction pipelining, caching algorithms, and probably several other areas which combine to make the same frequency CPU capable of 10 times the performance.
On this site, you can find several examples of older CPUs that have 1/10th, 1/20th or less the performance of a current model with the same frequency. As a really off the wall example, the Intel Xeon E5-2699 v3 @ 2.30GHz (Q2 2014) has almost 100 times the performance of the Intel Xeon 2.40GHz (Q1 2009).

Wrong part, perhaps?

You are comparing a dual core CPU to a quad core CPU, and the test at cpubenchmark.net is highly multithreaded. Even so, the Micro-6200T is only about 10% slower, which implies a much better IPC. Here is a better comparison. The Micro-6400T is optional on this mini PC according to TFA and is a quad core part. It performs about 50% faster than the Atom and is 80% of the performance of a Celeron J1900 while using 40% of the power. That seems pretty good to me. Granted the new Broadwell Core Ms will be faster still with similar power usage, but those are also much more expensive.

Beaten by an 18 month old Intel part

And the Intel chip is cheaper (contra revenue and such).

For example, the AMD E1 Micro-6200T in the Fitlet-B:
http://www.cpubenchmark.net/cp...

I have an i5-4690 Which is 3.5Ghz with turbo up to 3.9Ghz, and 84W TDP.

i7-4771 is rated at 3.5Ghz base, turbo of 3.9Ghz, and 84WTDP.

These benchmarks show identical single thread performance:
http://cpuboss.com/cpus/Intel-...
https://www.cpubenchmark.net/s...
i7-4771-2,229
i5-4690-2,228

I will second Sager, I have been very pleased with my Sager NP8255-S (aka Clevo P157SM). I is going into its 2nd year now. I ended up choosing the Sager because:
1. strong i7 cpu
2. up to 32gb ram
3. supports four hard drives
Actually "four hard drives" for this model means 2 x 2.5" and 2 x m2 slots. Which is huge, compared to the alleged mainstream performance workstations like Dell's Precision line or HP's z-books or Lenovo's W-series.
I'm running 3 x 400gb ssds in a raid 0 and I find that disk-intensive workloads are pleasingly fast.
I am getting the following in PassMark's Performance Test 8.0:
overall disk score: 5,558
seq read: 715 MB/sec
seq write: 523 MB/sec
random rw: 300 MB/sec
(just for the record, I run regular backups because because of the potential fragility of raid-0).
Why not a 4-drive raid? I figured I'd save a 2.5" slot for a multi-terabyte disk some day for on-board archives once I fill up the ssd's. (And I still have the optical bay to drop a caddy in if I need more storage).
Until 8x pcie ssd devices are available in laptops, raiding SATA together seems like the best way to boost lugable disk performance.
It loosk like the NP9752-S is the current model of this machine.


Now... if if you're looking for insane power in a laptop form factor, take a look at Eurocom's Panther.
If you really need crazy CPU cycles, this seems like a good choice:
PassMark for xeon E5-2687W v2

Here are the specs; I didn't go with the Panther because the cost-curve didn't work for me (money actually is an object in my case).
All-in-One Server with XEON 12-cores/24-threads, integrated display, keyboard and built in UPS (Uninterrupted Power Supply)
WEIGHT/DIMENSIONS: 5.5kg (12.1lbs); 419(W)x286(D)x57.9-62.1(H)mm (16.76x11.44x2.31-2.48inch)
SECURITY: TPM 1.2; Fingerprint, Kensington Lock
OPERATING SYSTEMS: Microsoft: Server 2012, 2008R2; VMware, VMware ESXi; Linux; RedHat 6.4 Enterprise Server Edition
CORE LOGIC: Intel C600/X79 Express Chipset
PROCESSOR: 12-core, 10-core, 8-core and 6-core Intel XEON E5-2600 and E5-2600 v2 series; up to E5-2697 v2 (12-cores/24-threads); socket LGA2011
MEMORY: up to 32GB; DDR3-1333/1600/1866; four physical SODIMM sockets
EXPANSION: Built-in ExpressCard 34/54 slot (for optional Expansion box required for Dual/Quad Port or Fiber LAN Adapter for i.e. for VMware ESXi)
STORAGE: up to 8TB of storage with four physical HDD or SSD, RAID 0/1/5/10 support; SATA 6Gb/s
NETWORK:on-board 1Gbe LAN (Intel 82579V); 2nd or Dual-port LAN Adapter(s) available via ExpressCard slot or via external expansion Magma box
OPTICAL DRIVE BAY: DVD-RW or Blu Ray Burner or 4th Hard Drive
I/O PORTS: 3x USB 3.0; 2x USB 2.0; eSATA; Firewire-800 (TI XIO2221ZAY); DisplayPort v1.2; DVI-I (SL); HDMI 1.4a out w/HDCP; Headphone; Microphone; S/PDIF out; Line-in; RJ-45 / LAN

I'm not trying to be a troll or flame here, so please don't take it like that. But, you may be comparing AMD processors and Intel in the same price range. Because the highest passmark bench results for laptops certainly belong to the 4th generation Intel mobile processors.

You must also work in an area of the country that is either really ahead, or really behind. I'm Sr sysadmin for a medium sized company and I haven't encountered a single person - outside of a geologist or engineer that needs real power - who prefers a desktop to a laptop in many years.

The benchmarks say that the CPU of the entry-level late 2014 Mac Mini is only 3.8% slower than the entry-level late 2012 Mac Mini. However, the TDP is also 57.1% lower (from 35W to 15W).

But it's a desktop PC, sacrificing a bit of power for a lower TDP is great in a laptop but to go 2 years on a desktop and release a system with a decrease in performance with only a lower TDP to show for it is pretty lame.

Yep, you are correct on that one. Even the fastest dual cores of the new Mac Mini can't kick the old quad cores' ass.

The new Mac Mini is twice as slow as the late 2012 model. So fuck you Apple.

The benchmarks say that the CPU of the entry-level late 2014 Mac Mini is only 3.8% slower than the entry-level late 2012 Mac Mini. However, the TDP is also 57.1% lower (from 35W to 15W).

Direct link to the comparison of both CPUs mentioned.