The SilverStone FT04 merges innovative modular design with a with a newly designed Fortress shell. Much of the FT02’s great performance remains but a host of nagging flaws prevent the FT04 from being a truly worthy successor.
September 23, 2013 by Lawrence Lee
| Product | SilverStone Fortress FT04 Extended ATX Tower Case |
| Manufacturer | |
| Street Price | US$200~$230 |
Though SilverStone is known for cool, classic looking cases, it is one of the
most innovative brands in the market, readily adopting new ideas and often committing
to them quickly. Many of their competitors are gun shy about veering from the
mainstream paradigm. SilverStone has shown no such qualms and this attitude
has helped to create some memorable high-end cases offering unrivaled cooling
and acoustic performance.
SilverStone combined elements that a few others had experimented with — enormous fans, positive pressure, and a rotated motherboard tray — to create the landmark Raven RV01. Switching to a bottom to top airflow dynamic reduced noise by placing the fans further away from the user. Turning the motherboard 90 degrees allowed the giant fans to directly blow cool air through the GPU and CPU area and out the top. Dual video card configurations also benefited as there was no longer an upper card being bombarded by the heat rising from the card below. The Raven was also surprising for SilverStone as it was dressed in a fanciful plastic exterior. Critics of this design were silenced when the second iteration RV02 was released alongside a more SilverStone-like aluminum twin, the Fortress FT02. These two models are the best performing cases we’ve tested.
|
|
Interestingly, the FT02’s successor, the FT04 (the FT03 was a much different microATX chassis) features a completely different design. This is evidenced by the dimensions — the FT02 is almost comically deep while the FT04 is much shallower and taller, with a footprint far closer to that of a “standard” tower case. It also has a softer, more contoured look and a thick concave front door. The bezel and intake vents are made of plastic, which purists will condemn as heresy. The door at least, is aluminum, and helps block out noise from the fans which have been moved to the front of the case. The rotated motherboard design has been taken one turn further; that is to say it’s completely upside down. These two features and indeed most of the internal layout is taken from the equally innovative SilverStone Temjin TJ08-E microATX case. The TJ08-E is a fairly cramped case, though, and many of the design choices were made to improve convenience of assembly and maintenance rather than for performance concerns.
The latest Fortress is effectively modular, with a removable top panel, motherboard tray, and hard drive cage. More drives can be crammed inside: 7 x 3.5 inch and 4 x 2.5 inch drives are supported. On the down side, the FT04 doesn’t offer as much cooling as its predecessor, having two rather than three 18 cm fans, and the 12 cm exhaust fan is optional, not included. The fans you do get are more powerful though, with a top speed of 2000 RPM. To tame these beasts, fan control is improved; both stock fans are connected to controllers with adjustable dials at the front of the case. We’re glad to see the speed switches all the way at the back on previous iterations have been deposed. Plastic support structures for graphics cards and the CPU cooler are also present. These are not features users have been screaming for, but they come in handy with heavy components.
The FT04 is available in four different variants, black and silver, with or without a window. The FT04 also has a twin in the Raven RV04, which uses essentially the same design with a plastic fascia and lower asking price. The two cases can be purchased for approximately US$230 and US$160 respectively.
|
|
|
|
The accessory box contains the obligatory assembly guide, bag of screws and standoffs, and a few zip-ties to help with cable management. Also included are brackets to mount up to a 3 x 12 cm radiator in place of the two 18 cm stock fans, and the VGA holder, a plastic contraption with tabs to physically support up to three graphics cards, and a pair of odd looking 2-pin plugs. The included fans are 3-pin but use a 2-pin connector to hook up to the fan controller — these can be used as place-holders if you opt not to use it.
| Specifications: SilverStone Fortress FT04 (from the product web page) | ||
| Model No. | SST-FT04B (black) | |
| SST-FT04S (silver) | ||
| SST-FT04B-W (black + window) | ||
| SST-FT04S-W (silver + window) | ||
| Material | Aluminum door and top panel, steel body | |
| Motherboard | SSI-EEB, SSI-CEB, Extended ATX, ATX, Micro-ATX | |
| M / B Support | — | |
| Drive Bay | External | 5.25″ x 2 |
| Internal | 3.5″ x 7 (2 hot-swap) , 2.5” x 4 | |
| Cooling System | Front | 2 x AP182 180mm intake fan 500~2000rpm, 17~42 dBA (compatible with 3 x 120mm fan) |
| Rear | 1 x 120mm fan slot | |
| Side | — | |
| Top | — | |
| Bottom | — | |
| Internal | — | |
| Expansion Slot | 8 | |
| Front I/O Port | USB 3.0 x 2 audio x 1 MIC x 1 | |
| Power Supply | 1 x Optional standard PS2(ATX) no length limitation | |
| Expansion Card | Compatible up to 13.3” long, width restriction-6.69″ | |
| Limitation of CPU cooler | 165mm | |
| Limitation of PSU | — | |
| Net Weight | 11.5kg | |
| Dimension | 219mm (W) x 546mm (H) x 482mm (D), 57.6 liters | |
EXTERIOR
The SilverStone Fortress FT04’s body is constructed primarily of steel. The
exceptions are the door and top panel which are composed of aluminum, and the
front bezel which is plastic. It measures 21.9 x 54.6 x 48.2 cm or 8.6 x 21.5
x 19.0 inches (W x H x D) for a total case volume of 57.6 L, considerably smaller
than most high-end/enthusiasts cases, including the FT02 which comes in at about
65 L.
We noted some obvious fit and alignment issues on our sample, as you’ll see
in the photos and captions below. The cause of these atypical (for Silverstone)
fit issues appears to be the sheer complexity of the design and its large size.
It has more interlocking parts and panels than most, and with the big one-piece
front door/fascia, only a tiny misalignment is enough to translate to a fairly
visible mis-fit.
 The door is impressively thick and lined on the inside with sound-deadening foam, as is much of the case interior. Three magnets secure the door loosely; it swings out with the gentlest of nudges. The plastic bezel surrounds a pair of 5.25 inch bays, two fan control dials, and a large curved fine mesh dust filter servicing twin 18 cm fans. |
 The fan controller can run the stock fans between 500 and 2000 RPM but the knobs are tiny, making it difficult to fine tune. |
 USB 3.0 and audio ports and located on the right side with the power and HDD LEDs above it. The bright blue lighting is diffused by small windows at the top of the door so it’s not too distracting. Notice the gap between the door and top panel — it doesn’t close flush. |
 On the left side of the bezel are plastic power and reset buttons. They require some effort to depress and oddly springy. You can see from this angle, once again, that the door and top panel don’t line up properly. |
 The left side of the case is a solid panel while a large window can be found on the right. Cases that rely only on front intake fans usually have airflow issues but the huge side intake vents bypass that issue completely. They’re also angled toward the rear, directing less noise toward the user. The case has a dull matte finish which doesn’t pick up fingerprints as easily as the FT02. |
 The case features an upside down (compared to the norm) layout so the power supply is located at the top and the CPU area is near the case floor. The rear of the case is well ventilated and if the two monster intake fans are not enough, there’s an additional 120 mm exhaust fan placement. |
 The square rubber case feet at the front are too low. As a result, the door actually scrapes the ground and if you look carefully you can see scuff marks that appeared less than 10 minutes into photographing the case. |
 We were surprised to see more alignment issues at the bottom of the case. The side panels sit more or less flush with the top of the case but at the bottom, they bulge outward. |
INTERIOR
Despite its modular nature, the FT04’s internal construction feels solid all the way around though it’s not quite up to par with the FT02. The only thing that feels cheap or weak is the removable drive cage as it lacks structural support, making it easily prone to vibration.
 The upside down removable motherboard tray, removable drive cage, CPU support lever and power supply placement are all elements borrowed from the Temjin TJ08-E. They’ve added a second fan, and a couple of extra drive bays at the bottom, one of which has a SATA backplane to make it hotswappable. |
 One thing SilverStone always gets right is cable management. All the appropriate holes are present and multiple holds for twist/zip-ties are littered across the back of the motherboard tray. To remove the tray, a few screws have to be removed, and then the tray can be pushed toward the front of the case and taken out. |
 The side panels are 0.9 mm thick and composed of steel. The door, floor, ceiling, and side panels are all lined with 3 mm sheets of stiff but springy acoustic dampening foam. |
 To remove the top panel, remove the two screws at the back and pull. The panel, like the door, has a concave shape. Another removable dust filter has been placed here for the power supply fan. |
 The 5.25 inch drives and the power supply are dropped in from above. |
 Holes have been drilled into the frame of the FT04 for straight access to the screws holding down the expansion slot covers. |
 The CPU heatsink support lever is attached to a drive bay on the bottom of the case. It has a sliding system to adjust its position. |
 The fans are positioned to blow through the main drive cage and expansion slots. For extra long graphics cards, there are plastic rails in front of the top fan to help support them. |
ASSEMBLY
Assembling a system in the FT04 is fairly straight forward. Our usual ATX case test system was used: An Asus 790GX motherboard with Phenom II X4 955 CPU cooled by a ZEROtherm FZ120 heatsink with a Nexus 120 mm fan, a WD Caviar hard drive, a Cooler Master 700W modular power supply, and one and/or two Radeon HD 4870 graphics cards (see full system details on the next page).
 Our test board secured to the removable motherboard tray. |
 The drive bay with the backplane doesn’t have a proper release mechanism, so SilverStone added a ribbon to pull the drive free. |
 Our test system with two Radeon HD 4870’s installed. |
 Tidying up the cables was a non-issue. |
 There is ample space behind the motherboard tray for extra cabling, about 24 mm. |
 The main drive cage is lined with foam but it’s only secured by two screws on the right side and a thumbscrew at the top. There’s no frame or rails on the left so it’s completely unsupported on that side. Our test drive actually vibrated less when mounted into the hotswap bay underneath. |
 A plastic bar with adjustable pegs acts as a physical support so longer, heavier graphics cards don’t droop down. |
 In addition to long 3-pin cables, the stock fans have short 2-pin dongles to connect them to the built-in fan controller. This basically means you can’t connect any other fans to the controller without modification. |
TESTING
System Configuration:
- AMD Phenom II X4 955 Black Edition
processor – 3.2 GHz, 45nm, 125W TDP - ZEROtherm Zen FZ120
CPU cooler with Nexus 120mm fan - Asus M4A78T-E motherboard
– 790GX chipset - Crucial memory – 2x2GB, DDR3-1333
- ATI Radeon HD 4870 1GB
graphics card, 150W TDP - Western Digital Caviar Black
hard drive – 1TB, 7200 RPM, 32MB cache - Coolermaster
Silent Pro M700W – modular ATX power supply - Microsoft
Windows 7 operating system – Ultimate, 64-bit - ATI
Catalyst 11.1 graphics driver
Measurement and Analysis Tools
- CPUBurn
processor stress software. - FurMark
stability test to stress the integrated GPU. - GPU-Z to
monitor GPU temperatures and fan speed. - SpeedFan
to monitor system temperatures. - Seasonic
Power Angel AC power meter, used to measure the power consumption
of the system. - Custom-built, four-channel variable DC power supply, used to regulate
fan speeds during the test. - PC-based spectrum analyzer:
SpectraPlus with ACO Pacific mic and M-Audio digitalaudio interfaces. - Anechoic chamber
with ambient level of 11 dBA or lower - Various other tools for testing fans, as documented in our
standard fan testing methodology.
System temperatures and noise levels were recorded with SpeedFan and GPU-Z
at idle, and on load using CPUBurn (K7 setting) and FurMark, an OpenGL benchmarking
and stability testing utility.
Baseline Noise
The main noise source within the FT04 are its two high speed 3-pin 18 cm fans.
They’re rated for 2,000 RPM which is really fast for any size fan, let alone
18 cm monsters. Thankfully, each fan is connected to its own dedicated fan controller.
| Stock Fan Noise Level (2 x HD 4870 Configuration, Idle) | |
| Fan Speed | SPL @1m |
| Off | 22 dBA |
| 500 RPM | 23~24 dBA |
| 800 RPM | 29~30 dBA |
| 1000 RPM | 35 dBA |
| 1400 RPM | 43 dBA |
| 1850 RPM | 55 dBA |
| CPU fan set to 100% speed. Ambient temperature: 22°C. | |
Our CrossFireX HD 4870 configuration assembled and running idle measured just
22 dBA@1m with the system fans disabled. Turning the system fans on to the lowest
controllable speed of 500 RPM added only 1~2 dB on top of that. A typical 12
cm fan running at this speed would be inaudible but a larger model spinning
at the same rate is much noisier. At 800 RPM, the noise level almost hit 30
dBA@1m which is pretty loud by our standards. Anything about 1200 RPM seems
like overkill for almost any kind of system. The high top speed is unwarranted
given the size of the fan and the tremendous airflow they can generate.
 |
While it doesn’t take a lot of speed to make the fans noisy, they sound far
superior to the 18 cm Air Penetrators in previous iterations of the Raven and
Fortress series, which had a very clicky character. We noticed the same difference
with the Temjin TJ08-E’s intake fan so it is quite possible that vertical rather
than horizontal orientation is better for the fans’ acoustics. At 800 RPM, they
didn’t add much, if any, tonality to the overall noise profile.
TEST RESULTS: SINGLE RADEON HD 4870 CONFIGURATION
| System Measurements (HD 4870) | |||
| System State | Idle | CPU + GPU Load | |
| System Fan Speeds | 500 RPM | 800 RPM | |
| CPU Temp | 27°C | 41°C | 39°C |
| SB Temp | 49°C | 58°C | 56°C |
| HD Temp | 30°C | 31°C | 30°C |
| GPU Temp | 72°C | 85°C | 85°C |
| GPU Fan (auto) | 950 RPM | 1870 RPM | 1800 RPM |
| SPL@1m [right side] | 23 dBA [23 dBA] | 26 dBA [26~27dBA] | 30~31 dBA [31~32 dBA] |
| System Power (AC) | 119W | 308W | 308W |
| CPU fan set to 100% speed. Ambient temperature: 22°C. | |||
Our single video card test configuration proved to be little challenge for
the FT04. Sitting idle, system temperatures were relatively low but at 23 dBA@1m,
it wasn’t as quiet as we would have liked. With some cases we’ve seen 20 dBA@1m
or lower in this state but that wasn’t possible with the FT02’s gigantic fans
at minimum speed.
On full load, as one would expect, only the hard drive was immune to double-digit
temperature increases. The video card’s fan was spinning at 1870 RPM, which
of course was the main contributor to the modest overall noise level of 26 dBA@1m.
The stock fans are completely overkill for this setup — increasing their
speed to 800 RPM resulted in minor temperature reductions and allowed the GPU
fan to slow by only 70 RPM. The improvements were not worth the spike in noise
to past 30 dBA@1m.
Usually, we measure case noise with our mic positioned diagonally from the
left side panel as most manufacturers place windows on the left and presumably
people would want to look inside their systems from their seated position. As
the FT04 has a nonstandard layout with the window is on the right side and the
back of the motherboard tray facing on the left side (blocking out a lot of
the noise produced on the other side) we felt it wouldn’t be fair unless we
measured the noise output from the opposite side as well. There was no difference
when the system was idle but on load it was about half a decibel lower. The
sound emitted by the GPU fan was more pronounced from the right side.
 |
| HD 4870 Configuration Comparison (Load) | ||||
| Case | SilverStone Fortress FT02 | SilverStone Raven RV03 | SilverStone Fortress FT04 | Antec P280 |
| System Fan Speeds | top at 9V, bottoms @11V/low | top, bottoms @9V/low | fronts @ 500 RPM | top, rear, front @low |
| CPU Temp | 41°C | 39°C | 41°C | 45°C |
| SB Temp | 49°C | 51°C | 58°C | 52°C |
| HD Temp | 30°C | 36°C | 31°C | 28°C |
| GPU Temp | 85°C | 83°C | 85°C | 85°C |
| GPU Fan (auto) | 1600 RPM | 1680 RPM | 1870 RPM | 1950 RPM |
| SPL@1m [right side] | 25~26 dBA | 25~26 dBA | 26 dBA [26~27dBA] | 27 dBA |
| CPU fan set to 100% speed All temperature results adjusted to 22°C ambient. | ||||
While the FT04 is one of the best performing cases in our single GPU test,
the lack of a 120 mm exhaust fan and third 180 mm intake fan makes it difficult
for the FT04 to compete with its predecessor. The FT04’s best result was with
the stock fans running at minimum speed but in this state, it was slightly louder
than the FT02, and the Southbridge chip, which sits close to the video cards,
was considerably hotter.
TEST RESULTS: DUAL RADEON HD 4870 CONFIGURATION
| System Measurements (2 x HD 4870) | ||||
| System State | Idle | CPU + GPU Load | ||
| System Fan Speeds | 500 RPM | 800 RPM | 1000 RPM | |
| CPU Temp | 28°C | 42°C | 40°C | 38°C |
| SB Temp | 57°C | 69°C | 67°C | 63°C |
| HD Temp | 31°C | 30°C | 30°C | 28°C |
| GPU #1 Temp | 78°C | 88°C | 86°C | 85°C |
| GPU #1 Fan Speed (auto) | 1000 RPM | 2150 RPM | 1960 RPM | 1940 RPM |
| GPU #2 Temp | 64°C | 84°C | 84°C | 82°C |
| GPU #2 Fan Speed (auto) | 1040 RPM | 1930 RPM | 1800 RPM | 1770 RPM |
| SPL@1m [right side] | 23~24 dBA [24 dBA] | 30 dBA [30~31 dBA] | 31~32 dBA [32 dBA] | 35~36 dBA [36 dBA] |
| System Power (AC) | 167W | 521W | 518W | 519W |
| CPU fan set to 100% speed. Ambient temperature: 22°C. | ||||
Adding a second video card obviously made the system interior warmer. Sitting idle, the Southbridge heated up by an additional 8°C while the top graphics card ran 6°C hotter. That being said, there was hardly any acoustic difference.
On load with the stock fans at minimum speed, once again, the CPU, Southbridge
and GPU took a major thermal hit, while the GPU fans about doubled in speed.
While 30 dBA@1m is fairly loud, in this configuration, it’s an excellent result.
Increasing the intake fans to 800 RPM tacked on another 1~2 dB for an across
the board (except for the hard drive) temperature reduction of 2°C. It looks
like the high speed stock fans are overpowered even for our CrossFireX test
system.
 |
| CrossFireX Configuration Comparison (Load) | ||||
| Case | SilverStone Fortress FT02 | SilverStone Fortress FT04 | SilverStone Raven RV03 | In Win Dragon Rider |
| Fans Speeds | top at 9V, bottoms @11V/low | fronts at 500 RPM | top @12V, bottoms @9V/low | top, rear, sides @9V, front @5V |
| CPU Temp | 41°C | 42°C | 41°C | 35°C |
| SB Temp | 64°C | 69°C | 66°C | 46°C |
| HD Temp | 29°C | 30°C | 34°C | 31°C |
| GPU #1 Temp | 86°C | 88°C | 87°C | 85°C |
| GPU #1 Fan Speed (auto) | 1830 RPM | 2150 RPM | 2140 RPM | 1890 RPM |
| GPU #2 Temp | 84°C | 84°C | 81°C | 82°C |
| GPU #2 Fan Speed (auto) | 1650 RPM | 1930 RPM | 1820 RPM | 1680 RPM |
| SPL@1m [right side] | 28 dBA [28~29 dBA] | 30 dBA [30~31 dBA] | 31~32 dBA | 32 dBA |
| CPU fan set to 100% speed. All temperature results adjusted to 22°C ambient. | ||||
Again, the FT04 delivered the best cooling:noise ratio with the stock fans
running at 500 RPM but in this hotter, dual graphics card environment, performance
fell further behind its predecessor. The Southbridge temperature gap narrowed
but with higher GPU fan speeds, the overall noise output was 2 dB higher, whether
measured from the left or right side of the case.
AUDIO RECORDINGS
These recordings were made with a high resolution, lab quality, digital recording
system inside SPCR’s own 11 dBA ambient anechoic chamber, then converted to
LAME 128kbps encoded MP3s. We’ve listened long and hard to ensure there is no
audible degradation from the original WAV files to these MP3s. They represent
a quick snapshot of what we heard during the review.
Each recording starts with ambient noise, then 10 second segments of product
at various states. For the most realistic results,
set the volume so that the starting ambient level is just barely audible, then
don’t change the volume setting again while comparing all the sound files.
- SilverStone Fortress FT04 – Baseline – 2 x Radeon HD 4870 test system at 1m
— stock fans off (22 dBA@1m)
— stock fans at 500 RPM (23~24 dBA@1m)
— stock fans at 800 RPM (29~30 dBA@1m)
— stock fans at 1000 RPM (35 dBA@1m)
— stock fans at 1400 RPM (43 dBA@1m)
— stock fans at 1850 RPM (55 dBA@1m)
- SilverStone Fortress FT04 – Radeon HD 4870 test system at 1m
— idle, stock fans at 500 RPM (23 dBA@1m)
— load, stock fans at 500 RPM (26 dBA@1m)
— load, stock fans at 800 RPM
(30~31 dBA@1m)
- SilverStone Fortress FT04 – 2 x Radeon HD 4870 test system at 1m
— idle, stock fans at 500 RPM (23~24 dBA@1m)
— load, stock fans at 500 RPM (30 dBA@1m)
— load, stock fans at 800 RPM
(31~32 dBA@1m)
FINAL THOUGHTS
Following its predecessors from the Raven and Fortress series, the SilverStone
FT04 boasts excellent cooling/acoustic performance against more conventional
competitors. It’s not up to snuff with the FT02,
ostensibly because it lacks the third 18 cm intake fan, but this combined with
the switch to front intakes results in a smaller footprint. The new version
is taller but not nearly as deep. The fans also sound less clicky, presumably
due to their horizontal orientation, and the fan controller is a big improvement
over the FT02’s two speed switches located inconveniently at the rear of the
chassis. The FT04 also adds a couple of extra hard drive bays and a bracing
system to keep the graphics cards level.
While a Fortress in name, what SilverStone really did was make a larger Temjin
TJ08-E and transplant it into a Fortress body, presumably to make a
more user-friendly high-end tower. Borrowing the innovative modular design of
the TJ08-E was well-intentioned but also introduced some problems. The removable
hard drive cage is designed for convenience rather than stability or noise reduction.
It’s held on with just three screws and there is no support structure above
or to the side to brace it. The side-to-side motion generated by mechanical
hard drives shake the cage, and some of the vibration is passed on to the rest
of the case. It’s ironic that the HDD cage is lined with dampening material
as the two extra drive bays on the bottom of the case, which have bare metal
on metal contact, offer quieter operation. The removable motherboard tray is
a nice but unnecessary addition as the interior is spacious and once the drive
cage is removed there’s nothing in the way.
The most disappointing aspect of the FT04 is a combination of small annoyances.
The door doesn’t close soundly, whether it’s because the magnets are too weak,
the door is too heavy or parts assembled just a touch askew. If the case is
on its side, gravity alone will pop it open, as well as just casually brushing
up against it. The short case feet are a bigger oversight. The door sits so
low that it scrapes against the floor. The top panel is also noticeably out
of alignment with the contours of the door. Even the side panels are imperfect,
flush against the rest of the chassis at the top but bulging out at the bottom.
We could forgive one or two of these issues but all together, in a SilverStone
case no less, they are disappointing. Many SilverStone fans were disappointed
when the FT04 was unveiled due to high plastic content, which they felt cheapened
the premium look and feel of the Fortress series. It’s a trivial complaint compared
to the surprising lack of overall attention to detail.
Most retailers carrying the FT04 have it priced at about US$230 which
is close to the lowest price you’ll find for the FT02. One site, Antares
Pro, a subsidiary of AVADirect,
is currently selling it for US$200 though this is probably sale pricing.
Even with a $30 discount, we’d skip the FT04 in favor of the older model
unless case depth is an issue. The FT02 offers superior performance, better
build quality, and is leagues ahead in terms of fit and finish. The latest Fortress
introduces a couple of minor improvements but these are unfortunately overshadowed
by newly acquired undesirable qualities.
Our thanks to SilverStone for the Fortress FT04 case sample.
* * *
Articles of Related Interest
Nanoxia Deep Silence 1
BitFenix Ghost: Quiet ATX Chassis
SilverStone Temjin TJ04-E Evolution Case
SilverStone Precision PS07: Budget MicroATX Tower
Silverstone Fortress FT02 Revisited
SilverStone Temjin TJ08-E: MicroATX Evolved
* * *
