各款式 筆電收購 ASUS筆電師收購 ROG電競筆電收購 收購ACER筆電 收購Msi電競筆電 收購微軟SUR 看全文
筆電收購
Posted on
Review of the AMD Radeon RX 7900 XTX and XT: the XTX beats the Nvidia RTX 4080 for $200 less, except in power efficiency and ray tracing, and the XT comes close 二手筆電收購
二手筆電收購
The most significant is a new chiplet-based approach, similar in concept to the one AMD uses for its Ryzen CPUs. Rather than building the entire GPU die on one manufacturing process—increasing the die’s size and therefore the chances that some or all of it could be defective—AMD is building the main Navi 3 GPU die on a 5 nm TSMC manufacturing process and a series of smaller memory controller dies (MCDs) on a 6 nm process. These chips are all linked together with a high-speed interconnect, which AMD says can transfer data at speeds of up to 5.3 terabytes per second.
The main graphics compute die (GCD) contains most of the hardware you think of when you think of a GPU—compute units, shaders, ray-tracing hardware, the media encoding and decoding block, and display output. Both the 7900 XTX and XT use the same Navi 31 GCD, but the XTX runs at higher clocks and has more CUs and stream processors enabled. The XTX has 96 CUs and 6,144 stream processors, while the XT has 84 CUs and 5,376 stream processors. Both cards represent a jump up from the Navi 21 die used in the RX 6900 series, which maxed out at 80 CUs and 5,120 stream processors (and that’s before you account for other performance-boosting improvements).
RDNA 3 brings a little more of everything to AMD’s top-tier GPUs, from CU counts to memory bandwidth.
Credit:
AMD
RDNA 3 brings a little more of everything to AMD’s top-tier GPUs, from CU counts to memory bandwidth.
Credit:
AMD
The MCDs all include a single 64-bit memory controller and 16MB of AMD’s Infinity Cache, and they demonstrate the advantages of a chiplet-based approach. The 7900 XTX has a 384-bit-wide memory bus and 96MB of Infinity Cache, where the 7900 XT has a 320-bit bus and 80MB of cache; to accomplish this, all AMD has to do is remove an MCD. The exact same MCDs can be reused up and down the stack with all of the different RDNA 3 GCDs that AMD chooses to release, from low-end products with a single MCD up to midrange GPUs that use between two and four. Defects in MCD dies won’t require the larger, more complex GCDs to be thrown out or binned, and vice-versa.
In the realm of all-new features for RDNA 3, there are three things of note. First, the GPUs include new AI accelerators, which could be useful both for the plethora of AI-assisted content creation that has sprung up in the last year and for AI-assisted upscaling (if AMD chooses to implement it in some future version of its FSR upscaling algorithm; both DLSS and XeSS use AI for upscaling, but FSR 2.0 doesn’t).
Second, the video encoding and decoding block supports hardware-accelerated encoding for the AV1 video codec, just like the RTX 4000 series and Intel’s Arc GPUs. This should be useful for both content creators and streamers who want to stream higher-resolution video or video at the same resolution while using less bandwidth.
And third, the “Radiance Display Engine” adds DisplayPort 2.1 support to the GPUs. Monitors that take full advantage of DisplayPort 2.1’s extra bandwidth don’t really exist as of this writing, but when they do, RDNA 3 GPUs will be able to drive 4K displays at up to 480 Hz and 8K displays at up to 165 Hz.
The actual GPUs
AMD is using the physical size of the 7900 series, in light of the huge 4000-series GPUs.
AMD is using the physical size of the 7900 series, in light of the huge 4000-series GPUs.
AMD is using the physical size of these new GPUs as a selling point, as well as the fact that they use typical 8-pin power connectors (there have been jokes). We’ve only seen the first-party cards AMD is making and selling itself, but if their partners follow their lead, these GPUs will be notable for looking like a run-of-the-mill 2.5-slot GPU. Unlike the RTX 4090 and 4080, they’re not so huge that you’ll need to worry about whether they’ll fit in your case, and you won’t need to learn the ins and outs of a new power connector.
Like some of the 6000-series cards, the 7900 series is somewhat odd in that it includes a USB-C port instead of a third DisplayPort; a USB-C-to-DisplayPort or USB-C-to-HDMI adapter or cable will let you use it like you would a regular display output, but for the (still rare-ish) gaming monitors with USB hubs and USB-C ports, it can handle display output and USB data over one cable. The 二手筆電收購gigabyte M28U I use does meet these criteria, so I appreciate the addition; for people who need three DisplayPorts, it means another adapter or a cable with different connectors at both ends.
The 2.5-slot 7900 series is still thicker than 2-slot cards like the Radeon RX 6800 (top).
Credit:
Andrew Cunningham
The 2.5-slot 7900 series is still thicker than 2-slot cards like the Radeon RX 6800 (top).
Credit:
Andrew Cunningham
The 7900 XTX is, by a very small amount, the larger of these two cards, though they’re in the same ballpark. The XTX is also the only one of the two with any kind of LED lighting—some relatively low-key white strips on the fan-side of the card, similar to what Nvidia is doing with the RTX 4090 and 4080. The XT does away with LED lighting altogether. If this is the kind of thing you’ll like, you’ll need to look at partner cards to see whose gigantic rainbow lights can brighten your case the most.
Game performance
Gaming testbed
CPU
AMD Ryzen 7 5800X3D (provided by AMD)
Motherboard
Asus ROG Crosshair VIII Dark Hero (provided by ASUS)
RAM
64GB DDR4-3200 (provided by Crucial)
SSD
Western Digital Black SN850 1TB (provided by Western Digital)
Power supply
EVGA Supernova 850 P6 (provided by EVGA)
CPU cooler
280 mm Corsair iCure H115i Elite Capellix AIO
Case
Lian Li O11 Air Mini
We don’t have a 6900 XT or 6950 XT to test against, so we’ll focus mostly on comparisons to Nvidia’s products (we have provided AMD’s own numbers for reference). The RTX 4080 is the most important comparison—it’s what the 7900 series has been priced in response to, and it’s the cheapest next-generation GPU Nvidia has on the market while we wait on the rebranded 12GB RTX 4080 to come out.
But we’ve also tested it against Nvidia’s RTX 3080 Ti, which is still being sold new for as much or more than AMD is asking for the 7900 series cards. Used and refurbished RTX 3080 Tis offer a better value, of course, but they come with the standard caveats: You don’t know where a used GPU has been, you have to be careful not to get scammed, and you usually don’t have a warranty to fall back on if something goes wrong.
Credit:
Andrew Cunningham
Credit:
Andrew Cunningham
We don’t spend a ton of time poring over synthetic benchmarks in GPU reviews, because their results aren’t always borne out in real-world games. But in this case, the results do capture the RX 7900 XTX and XT’s performance in a microcosm. The XTX is faster than the 4080 in the DirectX12-based Time Spy test and DirectX11-based Fire Strike test, while the XT runs a bit behind the 4080. But the 4080 does better in any test that touches ray-tracing, including the Port Royal benchmark and the ray-tracing-specific test. Both cards outrun the 3080 Ti in most tests, but it’s closer in the ray-tracing-enabled benchmarks.
Results with an asterisk were collected using a different test PC. Results should be comparable, overall.
Andrew Cunningham
Results with an asterisk were collected using a different test PC. Results should be comparable, overall.
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
AMD’s testing data comparing the RX 7900 XTX and XT to the 6950 XT.
AMD
Both cards acquit themselves well in actual games, too. In titles with no ray-tracing features included or enabled, there’s not a single one of our tests where the RTX 4080 can beat the RX 7900 XTX. The 7900 XT is usually behind the RTX 4080, but it generally keeps it close.
It’s true that AMD’s best card can’t catch up to the RTX 4090, which beats it across the board (and sometimes by a lot). If you’re solely focused on “performance crown”-type bragging rights, Nvidia is still the way to go. But that card is also $500 more expensive, so there’s absolutely no question which GPU offers the better value for the money.
So far, so good for the RX 7900 cards. But things get more complicated when you start tracing rays.
Hit-and-miss ray-tracing
Ray-tracing performance was one of the RX 6000 series’ biggest downsides. The situation does improve with the 7900 series, insofar as more ray-traced games will be playable on these cards than were playable on 6900-series GPUs. But there’s still a notable performance gap between both 7900 GPUs and the 4080. There are still some games, both in our tests and in AMD’s reference scores, where the 7900 XTX can score small wins. But the consistent wins for the 7900 XTX and near-misses from the 7900 XT that we’ve seen in other games don’t really materialize here.
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
Andrew Cunningham
AMD’s own numbers show similar results to ours: it’s not impossible for the RX 7900 XTX to beat the RTX 4080 in games with ray-tracing enabled, but those results are the exception and not the norm.
Both 7900 cards can handle ray-tracing at 4K reasonably well, especially in games where it’s used more sparingly (the sun-drenched glassy locales in Hitman 3 tax the card more than brakelight effects in Forza 5 or torches in Tomb Raider do). But in Quake II RTX and Hitman 3, the cards both struggle at 4K, and FSR 2.0 at its highest-quality setting doesn’t help the Radeon cards as much as DLSS can help Nvidia’s; the 7900 XTX performs a bit more like the 3080 Ti, not the 4080, in these ray-traced games.
The 1 percent low scores from Hitman are telling—the RTX 4080 with DLSS is twice as fast here as the RX 7900 XTX with FSR 2.0, which means that the game feels playable throughout on the RTX 4080, while the Radeon cards still struggle with choppiness.
This isn’t a disaster for AMD. Some of the games we tested ran quite well at 4K, and more challenging games like Hitman and Quake II RTX run pretty well if you step down to 1440p (or use a high-performance FSR setting, instead of the high-quality setting we default to for testing). But it does complicate the narrative a bit. The 7900 XTX and XT are good 4K graphics cards unless you need consistently great ray-tracing performance, in which case Nvidia still provides the better experience.
Good and bad news on power efficiency
The RX 7900 series still uses 8-pin power connectors, the same as older GPUs.
Credit:
Andrew Cunningham
The RX 7900 series still uses 8-pin power connectors, the same as older GPUs.
Credit:
Andrew Cunningham
AMD rates the 7900 XTX’s total board power at 355 W and the 7900 XT’s at 315 W (up from the 300 W it was originally announced at). The RTX 4080’s total board power is listed at 320 W, and the 3080 Ti’s is 350 W.
We measured average GPU power draw while running the Hitman 3 and Borderlands 3 benchmarks, just to get a basic sense of each card’s power efficiency.
There’s good and bad news for AMD here. On the bad side, the RTX 4080 is the more efficient card, especially in games with ray-tracing effects enabled—its average power consumption also came in quite a bit under the rated 320 W. The 7900 XT consumes a little more power to provide slightly worse performance in Borderlands 3, and the Hitman comparison is particularly unflattering.
Credit:
Andrew Cunningham
Credit:
Andrew Cunningham
The good news is that the 7900 XTX at least manages to provide better performance in exchange for its power use, as long as ray-tracing isn’t involved. And both cards are a major improvement over the power efficiency of the RTX 3080 Ti in non-ray-traced games, providing better performance in Borderlands 3 at similar power levels (for the XTX) and much lower power levels (for the XT). For most people, these cards should represent an efficiency upgrade even if they can’t quite catch Nvidia’s.
The smaller cooling systems on these cards plus the higher power consumption do mean that these Radeon cards run a bit hotter than the RTX 4080. While HWInfo measured its average temperature at 57.6 degrees during the Hitman test and 56.1 during the Borderlands 3 test, the 7900 XTX averaged 70 degrees in both tests, and the 7900 XT was 66.4 and 64.6 degrees, respectively. These aren’t dangerous temperatures, but especially in the XTX’s case, I could feel that extra heat being vented out of my case during extended gaming and benchmarking sessions.
Status quo
Xbox Series S for scale. The 7900 XTX is smaller than the 4080, but that doesn’t make it tiny.
Credit:
Andrew Cunningham
Xbox Series S for scale. The 7900 XTX is smaller than the 4080, but that doesn’t make it tiny.
Credit:
Andrew Cunningham
The Radeon RX 7900 XTX and XT don’t alter the balance of power in the GPU business. As has been the case for years, Nvidia has the edge in power efficiency, high-end performance, and add-on features like DLSS, while AMD gets by with pretty good cards that emphasize raw rasterized gaming performance for the money. At $899 and $999, these are a far cry from the RX 480 and RX 580 days a few years ago, but the dynamic is the same—AMD has undercut the RTX 4080 on price while staying within spitting distance on performance in most non-ray-tracing games (the 7900 XTX usually beats the 4080 in games we tested, while the XT trails a little behind).
Impressivel
（圖片來源／tweaktown）
AMD（超微）與 Intel（英特爾）兩家頂級處理器都已經準備釋出，旗艦款跑分、測試陸續出爐。根據《3DMark》資料顯示，AMD 新一代 Ryzen 9 3950X 處理器測試結果完全勝過 Intel 新款的 Cascade Lake-X 旗艦 Core i9-10980XE。
透過兩位《3DMark》用戶提供的跑分資料，AMD Ryzen 9 3950X 在 X570 二手筆電收購gigabyte Aorus Elite 平台上獲得 32082 分，處理器超頻最高超過 3,700MHz（其他核心為 3,500MHz）。而 Intel 的 Core i9-10980XE 則在華碩（ASUS）WS X299 SAGE 平台獲得 25,838 分，處理器啟用 Turbo Boost 最高達到 4,800MHz（基本時脈為 3,000MHz）。
（圖片來源／tweaktown）
從數據結果來看，Intel 在新一代旗艦處理器戰爭中處於下風，測試結果顯示 AMD 的 Ryzen 9 3950X 雖然採用的是 16 核心 32 線程，低於 Intel Core i9-10980XE 的 18 核心 36 線程，但在架構、製程勝於 Intel 的狀態下，AMD Ryzen 9 3950X 在算力與功耗表現均遠勝 Intel 旗艦，這是自 AMD 推出三代 Ryzen 後與 Intel 九、十代處理器差距最大的一次。
（圖片來源／tweaktown）
外媒《Tweaktown》指出，AMD Ryzen 9 3950X 目前仍屬測試階段，如果採用更好的超頻記憶體（3600~3800 MHz）做跑分運算，估計可以再提升 8～12％ 的成績，雖然對手 i9-10980XE 使用的測試平台也不是最頂級的，但雙方初次測試的成績就已經有顯著的差距，顯示 AMD 近年在處理器架構上做出的努力已經逐漸反映在效能上，反觀 Intel 長年採用 14nm 製程，處理器架構也沒有太大的改變，即使使用更多核心與功耗，表現仍不如對手。
外媒認為 AMD 已經在次世代處理器的戰爭中走在正確的道路，如果性能持續發展，新的 Threadripper 晶片將會成為新一代的旗艦王者。
【你可能還想看】
二手筆電收購 二手筆電收購