Saber TyT Hardware y Software common mistakes
Saber TyT Hardware y Software · Understand the traps and how to avoid them.
Memorizing component names (CPU, RAM, HDD, PSU) without being able to state the function, the unit of measure, or the connector/interface each one uses, so scenario questions about which part to replace or upgrade are answered by name association alone.
Candidates study a parts list as vocabulary, often from a single summary sheet, and never build the mental model of how data flows between CPU, cache, RAM, storage and bus, so they cannot reason about symptoms.
What to do instead
For each component write a three-line card: function, unit/speed characteristic (GHz, GB, RPM, W, MB/s), and interface or socket (SATA, NVMe M.2, DDR4 DIMM, PCIe x16, ATX 24-pin). Then practise mapping a described symptom to the component that explains it, justifying the choice in one sentence.
Treating preventive maintenance as a single task ('clean the PC') instead of a scheduled set of differentiated activities per component, and confusing preventive with corrective actions in scenario items.
Study material usually lists maintenance as a bullet list without timeframes or acceptance criteria, so candidates cannot distinguish cleaning a heatsink from replacing a failed fan, or a scheduled backup from restoring data after a crash.
What to do instead
Build a maintenance calendar table with columns: component, action, frequency, tool/consumable, risk if skipped. Include at least PSU, fans/heatsinks, thermal paste, storage SMART check, UPS battery, and backup verification. For each row, classify it explicitly as preventive or corrective and practise sorting given actions into the two columns.
Answering operating-system questions by brand familiarity rather than by function, e.g. failing to distinguish kernel, driver, file system, and user interface, or misclassifying a utility as an application.
Candidates use Windows daily but have never separated the layers, so they judge by what they see on screen and treat the GUI as the operating system itself.
What to do instead
Draw a layered diagram (hardware, kernel, drivers/HAL, services, shell/GUI, applications) and place ten named examples on it. Then answer practice items by naming the layer involved before choosing an option, and note which layer a driver update or a service restart actually touches.
Confusing software license categories, assuming freeware is open source, that a site license allows unlimited installation, or that an OEM license can be moved to new hardware.
The terms overlap in everyday use and candidates rely on intuition about price rather than on the rights granted: use, copy, modify, redistribute, transfer.
What to do instead
Make a comparison grid of proprietary/commercial, OEM, volume/site, freeware, shareware, open source (with GPL versus permissive distinction), and subscription. For each, fill in: cost model, right to modify, right to redistribute, transferability to another machine. Then resolve five short licensing scenarios against the grid.
Mixing up network topologies and their failure behaviour, e.g. claiming a star network collapses when one cable fails, or that a bus topology needs a central switch.
Diagrams are memorized visually without the underlying rule about shared medium versus point-to-point links, so the consequences of a single fault are guessed.
What to do instead
For each topology (bus, star, ring, mesh, tree, hybrid) state three facts: how nodes connect, what happens when one link or one node fails, and cost/scalability trade-off. Then answer fault-scenario items by first identifying whether the medium is shared or dedicated.
Confusing the OSI/TCP-IP layers with devices and protocols, placing a switch at layer 3 or a router at layer 2, or assigning HTTP to transport layer.
Layering is studied as a list of seven names rather than as encapsulation, and device behaviour (MAC versus IP forwarding) is not linked to the layer model.
What to do instead
Create a mapping table: layer number, PDU name, addressing used, example protocols, device operating there. Then trace one web request step by step from application down to physical and back, naming the PDU and address at each layer. Repeat with a ping to reinforce ICMP versus TCP/UDP.
Mixing up IPv4 addressing concepts: private versus public ranges, default mask by class, network versus broadcast address, and when a gateway or DHCP is required.
Candidates memorize '192.168 is private' but never practise subnetting arithmetic or the purpose of each configuration field, so items about host counts or unreachable hosts are answered by pattern matching.
What to do instead
Practise ten short exercises: given an address and mask, compute network address, broadcast, usable host range and host count. Separately write one line on the role of IP address, mask, gateway and DNS in a client configuration, and test which field is at fault in described connectivity failures.
Classifying threats only by name (virus, worm, Trojan, phishing, ransomware) without linking each to its propagation method, typical symptom, and the matching countermeasure.
Lists of malware types are memorized in isolation, and social-engineering attacks are treated as a separate topic rather than as an entry vector for the same payloads.
What to do instead
Build a threat matrix with columns: propagation vector, what it modifies, visible symptom, primary control (antivirus, patching, MFA, awareness training, backup, firewall). Fill it for at least eight threats, then work scenario items by identifying the vector first and the control second.
Giving generic security advice ('use strong passwords', 'be careful with email') instead of the specific practice tied to the risk, such as 3-2-1 backups, least privilege, patching cadence, or verifying sender domains.
Exam items often ask for the best control in a situation, and candidates default to the most familiar slogan rather than the control that addresses the stated risk.
What to do instead
For each risk scenario, write the control, the layer it acts on (human, endpoint, network, data) and one measurable indicator (e.g. backup restore tested monthly, MFA on all remote access). Practise choosing between two plausible controls by asking which one breaks the attack chain described.
Diagnosing hardware faults by jumping to the most expensive component, or skipping the ordered troubleshooting method (identify symptom, establish theory, test, verify, document) in favour of guessing.
Real-world experience is often limited to swapping parts, and exam scenarios reward a documented sequence plus the cheapest discriminating test first, which candidates skip under time pressure.
What to do instead
Memorize and apply the ordered method to five written cases: no POST, beep codes, random reboots, overheating, no display. For each, list tests in ascending cost order (visual inspection, reseat, swap cable, swap RAM, test PSU with multimeter, substitute known-good part) and state the evidence that confirms the diagnosis before replacing anything.