NVIDIA-NVIDIA,GeForce GTX GeForce GTX 680 preview 680,GK104,Kepler preview
Following After yesterday 3DCenter.org, the BSN website has always been good at depth analysis also released today the news about NVIDIA 28nm process Kepler architecture GK104 core: part of the site yesterday reported substantially the same, but there are some nuances worth pondering. (On yesterday's rumors PCINLIFE administrator cho's uncharacteristically did not throw a \"Fake\" or \"makes up\" left, in addition to a variety of sources has been indirectly implied the majority of the details are true)
First the core frequency and Shader frequencies on NVIDIA unified this long reports. Kepler framework, a set of SM CUDA Core amount of contrast Fermi structure GF114 was promoted to double, from 48 to 96. Memory controller for the 256bit interface, these yesterday 3DCenter message.
Next, let's look at the core schematic of GK104:
Big eyes and a look the and GF110 similar, but a closer look can be found in the SP number jumped from 512 to 1536. BSN said initially the NVIDIA consider the GK104 listed early name is the GeForce GTX 660, but 99% of the officially listed the name of GeForce GTX 680, the remaining 1% for the GTX 660/670. Price and before different mouth Charlie said: GeForce GTX 680 located in the GeForce GTX 560 Ti 2GB/GeForce GTX 560 Ti 448 Core / GeForce GTX 570's replacement, the initial price of 349-399 dollars. The performance is certainly better than the previous generation flagship GeForce GTX 580 to be much higher, the most important thing is the GK104 able to not only get rid of $ 449 Radeon HD 7950 3GB, $ 549 Radeon HD 7 970 also mention the news that 3DMark 11 X-mode score may about 3000, the night blast down 1,000 yuan seems about to appear ... (Hint: 1536/1024 or 768/512)
Next is the framework for analysis of GK104, Kepler, times a set of SM 96 CUDA Core, a GPC (Graphics Processing Clusters) containing 4 SM 384 CUDA the Core. The GK104 containing 1536 (4GPC), 2304 CUDA the Core is not difficult to speculate top GK110 may contain 6GPC. The memory controller may still be the low end entry level 64bit, midrange mainstream 128bit, high-end 256bit the top 384/512bit-- should be used in HPC, or a GK110 the GeForce GTX 690. Kepler's mobile graphics core is already long been taped to complete CES, Samsung has demonstrated a laptop equipped with the most low-end (96CUDA the Core). (Driver identify frequency / specifications)
Tahiti, XT, contrast and AMD GK104 on the number of ROP and TMU - are 32 and 128, but the memory controller indicators continued the \"traditional\" as good as AMD, 2GB 256bit GDDR5 memory on on Raeon HD 7 970/7950 of 3GB. However, the advantage on price and performance I think we will also force the rival product price.
And GF110, GK104 fact, for half of the single-precision, double precision floating point performance in two versions - Production Quadro and Tesla, the theoretical production GeForce quarter.
Comparison with the the GK104 core frequency is tentatively set at 950MHz, BSN, said the sources said that NVIDIA's laboratory initially available in two versions: 772MHz for the same the GeForce GTX 580, 925MHz for the same Tahiti, XT, for comparison, the actual shipping version of the core frequency may be between 950-1000MHz. Of course, the little blue pill for like AMD, NVIDIA is also stepping up to adjust the BIOS, fight for the public version of the the GK104 core frequency will stand on a 1GHz. If NVIDIA can not accomplish this goal, it will be the first time open to non-public version / AIC overclocked version.
Memory area, a tentative frequency of 1.25GHz equivalent 5GHz, relatively GF100/110 of up to 25%. This regard, AMD, GDDR5 memory / controller to be regarded as the standard-setting, one of AMD's GPU engineers to improve the GDDR5 memory has advantages also not unusual. Calculated by the previously described frequency, GK104 with 160GB / s memory bandwidth, compared to the GTX 580's 192.4GB / s, a slight decrease compared to the expected substitution products GTX, 560 of Ti and GTX, 560 Ti the GF110 core version / GTX, 570 128.27GB / s and 152GB / s have raised.
All these specifications, the single-precision floating-point capability 2.9-3.05TFLOPS, double-precision floating-point for the 486-500GFLOPS. Version of Quadro and Tesla will have to unlock the double-precision floating-point computing power - that is, around 1.5-1.6TFLOPS and GTX 580 single-precision floating point capability is quite significantly higher than the Fermi core Quadro and Tesla.
BSN concluded by stating that users will not wait too long, NVIDIA is poised to start Kepler's pre-sale / Distribution operations (including mid-GK107 - GeForce GTX 650, etc.), used against AMD will market action February 15 release of \"Cape Verde\", CeBIT in Hanover, Germany, the Radeon HD 7700 series, and in March, posted on Pitcairn Radeon HD 7800 series. Perhaps next month on the market will be able to see Kepler entity.
GK104 will be named GeForce GTX 670 Ti? -NVIDIA, Kepler, GK104, GTX 670 Ti?
GeForce GTX 560 Ti to achieve a \"titanium\" signs of a successful revival of the word, but later some of the GeForce GTX 550 Ti unworthy of the name, and the latest from the domestic hardware forum Bao Liu said, NVIDIA next-generation architecture of the end of Kepler core GK104 will be named \"GeForce GTX 670 Ti\".
GK104 is now generally believed that the upgrading of GF114 core, a series of performance-level market, according to NVIDIA's product naming in recent years, code-named the last one is 4 core tend to be x60, such as GeForce GTX 460, GeForce GTX 560 / Ti, etc., will be different this do? Was later increased GeForce GTX 660 Ti, GeForce GTX 660? These currently unknown, but it looks like NVIDIA seems to be more \"Ti\" suffix of the signs.
In addition, according to the OBR has always been more reliable reports, Kepler's first product will be released in April, is probably the GK104.
Yesterday we just said, NVIDIA has started shipping to notebook manufacturers Kepler chip, this will be the new Intel Ivy Bridge in April along with the market, maybe that is the core of the derivative GK104.
Thank Zhang Qiang provide news clues
Alien to mini machine:-Dell Alienware X51 games host, DELL,Alienware X51, mini, NVIDIA,Optimus
Alienware's gaming machines in the brand image is more deeply rooted. Just this afternoon, Dell held a conference in San Francisco, California, announced the latest Alienware products - the brand's first mini X51. Overall size and the old version of the Xbox 360 65nm process almost.
Although the size small, but Alienware is this product for gamers since it naturally will not use Atom / Brazos APU platforms. X51 uses an H61 mini-ITX motherboard, use the desktop version of the LGA 1155 Core i3/i5/i7 processor, providing maximum 8GB DDR3 memory (standard 4GB). Motherboard supports USB 3.0, integrated WiFi module and a slim DVD burner / Blu-ray drive.
The most important graphics, X51 has a PCI-Express x16 slot, support up to 9 inches around 228.6mm of the dual-slot graphics card. Hard drive cage to support installation of a 3.5-inch hard drive or a pair of 2.5-inch HDD (but need to transfer), can also use the 3.5 in. HDD +2.5 inch SSD combination. In addition, X51 and M17x/M18x notebook also has a similar external power supply module.
Specific configuration section, Alienware X51 standard graphics for the GeForce GT 545/GeForce GTX 555, both of which are OEM-specific models, with GDDR5 memory. GT 545 is a GTX 550 Ti shrink version, CUDA cores down to 144, down to 870 core frequency, memory frequency is equivalent to about 4GHz, width is 128bit. The GTX 555 is the GTX 560's scaled-down, similar to the GTX 460 SE specification and GTX 460 V2 combination, with the former and the latter the number of CUDA cores and other specifications: 288 CUDA Core, 776MHz core frequency, 192bit memory interface , equivalent to about 4GHz memory frequency. Most importantly, both the desktop graphics card also supports NVIDIA's Optimus technology, can greatly reduce the use of power.
General configuration details are listed below:
-CPU: Intel Core i3 (standard) dual-core / i5 or i7 quad-core processor
- Motherboard: H61 chipset Mini-ITX specification
- Graphics: NVIDIA GeForce GT 545 1GB GDDR5 (standard) / NVIDIA GeForce GTX 555 1GB GDDR5, Optimus intelligent switching technology to support
- Memory: 4GB DDR3-1333 (standard), upgradeable to 8GB, 2 DIMM slots
- Hard disk: Standard high-speed 7200rpm 1TB (picture point of view seems to be Western Digital Blue plate)
- Optical Drive: Slim DVD burner (standard) / Blu-ray
- Expansion: Gigabit networks, 802.11 a / b / g / n WiFi, 6 个 USB 2.0,2 a USB 3.0, HDMI 1.4a output, SPDIF output, HD audio output
- Power Supply: Alienware custom 240W (standard) / 330W
- Size: 343 * 95 * 318mm
- Weight: about 5.49kg
The Dell price is relatively kind, priced at just $ 699, the high-priced strategy for the Alienware is consistently regarded as valuable, and then take into account the work problem, some users considered a good choice.
NVIDIA Kepler named to determine the core graphics-NVIDIA, Kepler, GeForce GTX 680, GeForce GTX 550
Czech OBR today and new material can be excavated: first admitted his mistake: implications for Kepler to will be released in February, the thesis originated from old Roadmap last July, which means that Kepler was officially released in April. Though February is not a new card, is released GeForce GTX 550, from the model maybe is GTX 550Ti version 128bit or low-frequency version? Kepler published April specific model will also be the GeForce 600 instead of the 700 series.
NVIDIA GK104 renewed speculation: 32 million transistors $ 399-NVIDIA, GK104, rumors, specifications
Morning, the just-concluded NVIDIA CES keynote CEO Jen-Hsun Huang did not disclose on GeForce / Quadro / Tesla-related news. But it's not naturally there are other official sources, the Czech Republic today has released an OBR next generation NVIDIA core-side main GK104 Kepler detailed specifications chart.
As a fact, GeForce GTX 560 Ti's successor, GK104 integrating 3.2 billion transistors, design performance than the GeForce GTX 580 by about 10% -15%, depending on the frequency difference in AMD Tahiti or between family Radeon HD 7950/7970 and Radeon HD 7970 quite. Said to control power consumption is also good, the core can easily over the same frequency on 1100MHz. Pricing and GeForce GTX 570 is the equivalent of $ 399. Specifications described in the following figure, we can see that the previously reported core / Shader frequency of unity has been confirmed from multiple sources:
But on this memory controller is still in doubt, PCINLIFE administrator cho in the Radeon HD 7970 evaluation data should be implied 256bit. If you use 256bit memory controller, then NVIDIA is bound to push the memory frequency to 5.5GHz and AMD like to get around the memory bandwidth and reliable manner as not to become a bottleneck.
EVGA laboratory visits: four GTX 580, Dual LGA2011-EVGA, X79, GTX 580, Classified, Dual LGA2011, SR-X, k | ngp | n
Support two star LGA2011 Xeon E5 processor of SR-X double road Board, HPTX Super plate type two a processor socket and 8+4 section DDR3 memory slot pure Digital PWM powered circuit back I/O interface seven section PCI-E x16 slot, support PCI-E 3 and independent switch can separate opened and close each processor, and PCI-E slot, and SAS interface of switch switch silhouette [Shang a page][1th page][2nd page] click in this page reading full text
Leading up to apt-GTX580! HD7970-AMD,7970 overclocking experience, overclocking, 1.26GHz,6.9GHz, three screens
Radeon HD 7970 technology, performance, power consumption is very good, overclocking is the number one players, but now because of the lack of tools, but not very casually over too high. HardOCP recently did a special test, the Radeon HD 7970 for a more in-depth overclocking experience, the highest over to 1.26/6.9GHz, and had Eyefinity three-screen game testing.
How to overclock Radeon HD 7970 []
A card overclocking is by far the most convenient Catalyst Control Center OverDrive feature, you can directly drag the slider, but may be in order to ensure absolute safety, AMD has done a lot of limitations, core / memory frequency could only mention 1125/6300MHz, only higher than the original frequency 200/800MHz, obviously not fun.
Most third-party overclocking tools so that only MSI has just updated Afterburner can theoretically over after the break through to 1390/7160MHz, the actual test that can be pressurized to 1.3V to the core 1.2GHz ultra above.
Sapphire also provides an internal test version of the TRIXX Utility, can also unlock the Radeon HD 7970's overclocking limits. Through repeated testing, the final voltage to 1.3V, the core overclocked to 1260MHz, memory overclocked to 6900MHz, very stable!
Manual fan speed is fixed at the same time 75%, is quite noisy. The bottom of the power control settings both down-energy, to ensure the stability of the effect of overclocking, the overclocking can be pulled right +20%, thereby increasing control TDP.
Test platform configurations:
Processor: Intel Core i7-2600K OC 4.8GHz
Motherboard: ASUS P8P67 WS Revolution
Memory: Corsair DDR3-1600 8GB
Hard Drive: Western Digital 640GB Black Pad
Power Supply: Enermax MaxRevo 1350W
Display: Dell 3007WFP, Dell 2480WFP × 3 (Eyefinity)
Operating System: Windows 7 64 位 Ultimate
Graphics: Radeon HD 7970, GeForce GTX 580
Graphics Driver: AMD_Radeon_HD_7900_Win7_64_Dec16.exe (Dec20 actually better), GeForce 290.36
GALAXY GeForce GTX 580 is the version of the black four-star, three-screen support for 1080p output, the test overclocked to 860/4000MHz.
Test in a single screen resolution of 2560 × 1600, three-screen resolution of 5040 × 1050.
Continue>> Next [Page 1] [Page 2] [page 3] [page 4] [Next] Click to read full text in this page
AMD's official response: Radeon HD 7970 does not hide the SP-AMD, Radeon HD 7970, Tahiti XT, hide, SP
Sapphire last week leaked document exposed the existence of 40CU 2304SP Tahiti XT core \"full version\" of the rumors that AMD has not yet announced the \"Southern Islands\" series GPU architecture of the core thumbnails, so similar to the speculation can only temporarily on paper.
And good at digging deep behind the rumors of the Bright Side of News Site in the New Year holidays, contact the first time after the end of the AMD's senior public relations department spokesman Chris Hook and Eric Demers and get a direct response:
Chris Hook, said bluntly: \"Tahiti XT does not hide the core.\"
This once and for all - and unlike the Intel power consumption and heat cause castration as a Sandy Bridge-EP core and two 5MB L3 Cache makes Sandy Bridge-E; also like NVIDIA as the first initial yield factor launched the 480 CUDA Core of the GTX 480 instead of the full 512CUDA Core of the GTX 580. AMD 43 million transistors used in the full 384bit memory interface, 32 CU in 2048 on the SP, there is no space to place the current work flow processor that is hundreds of millions of transistors.
As for the current Radeon HD 7970 launched a test - pressurized or unpressurized overclocking, three / four-way exchange of fire, etc. has it all, who held the wrong message to the Sapphire is no need, Tahiti XT has proven it is a good core.
Of course, AMD does not rule out an official statement or with water - after all the data on the slide has been repeatedly proven to fake to the horizon.
AMD Radeon HD 7000 series price performance disclosure-AMD, Radeon, HD 7000
AMD preparing next month the latest “ southern islands ” series GPU products, on the basis of a series of desktop graphics products. Led will be code-named “ Tahiti ” (Tahiti), high-performance GPU, and the Radeon HD 7,900 series video cards. And then publish the code number “ Pitcairn ” (Pitcairn) Radeon HD7800 series of high performance GPU and a corresponding video, released after code “ Thames ” (Thames) and “ &rdquo on Lombok Island, (Lombok) mainstream-level GPU and corresponding Radeon
HD7700, Radeon HD7600 series.
Today, we are told that, Radeon HD7970 higher performance than GeForce GTX580,Radeon HD7970 graphics card is priced at $ 499, Radeon HD7950 higher performance than GeForce GTX570, priced at $ 399. &Amp;ldquo; Pitcairn ” (Pitcairn) high performance GPU will be Barts“ buzz ” successor.
This GPU is designed for the performance of a dessert-location design, specific models with Radeon HD7870 and HD7850, they will compete with the GeForce GTX560 Ti/GTX560, priced at US $ 299 and $ 199 respectively. Radeon HD and 7,670 respectively at $ 179 and $ 119. In addition, the Radeon HD and HD 7,970 7,950 allegedly will match 3GB graphics memory. According to the latest leaks news of Radeon HD7000 series and NVIDIA product performance are as follows: Radeon HD 6990>GeForce GTX 590>Radeon HD 7970>GeForce GTX 580>Radeon HD 7950>GeForce GTX 570>Radeon HD 6,970.
Kepler could not come from the external overclocking module Sotec play – Sauter, GTX560Ti, Extreme Edition, overclocking external modules, Kepler
Although the next-generation 28nm process, \"Kepler,\" the news of the outflow, even in the driver, but NVIDIA provides a road map from view, until after the Lunar New Year launch of the plan are not new, the graphics card manufacturers still need to continue to root out the GeForce 500 series potential, continue to give more.
Following the launch GTX570 Extreme Edition, ZOTAC once again supreme in its series on high-end roots, prepared a new version of GTX560Ti Extreme Edition, based on the original version in March of this year and make a newer, more sophisticated, luxurious and very original materials based on the cause of the upgrade again.
New PCB positive
Backside of the PCB version
Most notably, Sauter This time also joined the world's first external structure of M-PCIE overclocking module. The module is connected via USB cable, with the latest version of FireStorm tools, not only the freedom to pressure, but also see the power status of each phase, the entire card, power consumption, real-time data, one set of overclocking and intelligent monitoring.
It is reported that the external overclocking module is now in the measured phase will be formally shipping.
External overclocking module interface
External overclocking module PCB
FireStorm software
Materials used, the new GTX560Ti Extreme Edition also features enhanced IR Dr.Mos electrical performance management, and integration Sotec tailor AIO inductance, not only to optimize the thermal profile, but also has a fantastic sense of value. Capacitance in order to match the overall tone, into the Japanese black custom capacitors.
Comparison of old and new PCB
AIO inductance photograph
AIO inductance and capacitance of Japanese black
Back still equipped with a large number of MLCC capacitors to ensure stability of high-frequency current work. Super overclocking artifact FPCAP MLCC decoupling capacitors from an increase of two. This $ 10 an ordinary capacitive decoupling capacitors have unmatched capacity and performance.
Power supply is designed to follow the Sotec's D-Power VRM technology, using CHiL 8318 digital PWM chip design with mixed-mode analog circuits.
Also retained the Power Monitor power supply control circuit, to protect the safety of graphics performance. The PCB over the shoulder part of the voltage measurement is added six points, LN2_OTP liquid nitrogen overclocking mode limit switch.
Sauter GTX560Ti Extreme Edition was in June this year over the core frequency to crazy 1.67GHz, 3DMark Vantage scores also exceeded P35000, and this new version will be enhanced mid-December, the specific performance of the wait and see.
